diff --git a/00-ProjectProposal.ipynb b/00-ProjectProposal.ipynb
index 67ace3a..e7840dc 100644
--- a/00-ProjectProposal.ipynb
+++ b/00-ProjectProposal.ipynb
@@ -20,7 +20,7 @@
"\n",
"- Lui Gazem: Analysis, Writing – original draft\n",
"\n",
-"- Candance Eastep: Background research, Writing – original draft, Writing – review & editing\n",
+"- Cadence Eastep: Background research, Writing – original draft, Writing – review & editing\n",
"\n",
"- Ahgean Davis: Data curation, Software, Visualization\n",
"\n",
@@ -66,7 +66,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
-"Mobility rates and COVID-19 case counts exhibited a substantial correlation throughout the pandemic. In response to the rapid spread of the virus, public health authorities urged individuals to remain at home whenever possible, recognizing that increased movement and social interaction heightened the risk of transmission. From the onset of this global crisis, policymakers closely monitored the relationship between population mobility and infection rates in an effort to mitigate uncontrolled surges in cases. A study conducted by the Centers for Disease Control and Prevention (CDC) examined anonymized cell phone GPS data to assess patterns of movement across counties in the United States. The researchers observed that reductions in trips outside the home were associated with subsequent declines in reported COVID-19 cases. Specifically, changes in mobility were analyzed in relation to new case counts approximately 11 days later, reflecting the typical incubation period and reporting delays. The study also compared urban and rural areas to determine whether mobility patterns affected transmission differently across settings. This distinction was critical, as urban centers tend to be more densely populated, exhibit distinct transportation behaviors, and may experience differing timelines of viral introduction and spread compared to rural communities.\n",
+ "Mobility rates and COVID-19 case counts exhibited a substantial correlation throughout the pandemic. In response to the rapid spread of the virus, public health authorities urged individuals to remain at home whenever possible, recognizing that increased movement and social interaction heightened the risk of transmission. From the onset of this global crisis, policymakers closely monitored the relationship between population mobility and infection rates in an effort to mitigate uncontrolled surges in cases. A study conducted by the Centers for Disease Control and Prevention (CDC) examined anonymized cell phone GPS data to assess patterns of movement across counties in the United States. The researchers observed that reductions in trips outside the home were associated with subsequent declines in reported COVID-19 cases. Specifically, changes in mobility were analyzed in relation to new case counts approximately 11 days later, reflecting the typical incubation period and reporting delays. The study also compared urban and rural areas to determine whether mobility patterns affected transmission differently across settings. This distinction was critical, as urban centers tend to be more densely populated, exhibit distinct transportation behaviors, and may experience differing timelines of viral introduction and spread compared to rural communities.\n",
"\n",
"For our project, we aim to more closely examine variation in COVID-19 case counts across days of the week, with particular attention to identifying when case spikes are most pronounced. Consistent with our hypothesis, we anticipate that reported cases will be higher during weekdays compared to weekends. One possible explanation is that essential workers continued in-person employment throughout much of the pandemic, increasing opportunities for exposure during the workweek. Workplace environments may facilitate transmission due to repeated close interactions with coworkers and contact with shared surfaces (e.g., counters, equipment, communal appliances). These routine occupational exposures could contribute to elevated transmission risk during weekdays, which may subsequently be reflected in higher reported case counts. This hypothesis is supported by prior empirical work conducted by Hannah Hoffman and The COVID Tracking Project, which examined temporal reporting patterns in state-level COVID-19 case data. Their analysis demonstrated a consistent decline in reported case counts over weekends. Importantly, this pattern was attributed not only to behavioral changes but also to structural factors, including reduced testing, reporting delays, and public health regulations that limited activity during weekends in many states at the time.\n",
"\n",
@@ -90,6 +90,7 @@
"We hypothesize that daily COVID-19 case counts (daily_cases) will vary systematically across days of the week, with lower reported cases on weekends (is_weekend = 1) and higher reported cases during weekdays. Additionally, we predict that increases in workplace mobility (workplaces) will be positively associated with daily COVID-19 case counts, while increases in residential mobility (residential) will be negatively associated with daily case counts at the county level over time.\n"
]
},
+
{
"attachments": {},
"cell_type": "markdown",
@@ -246,7 +247,7 @@
"source": [
"## Team Expectations\n",
"\n",
-"Our team members are Nicholas Campos, Lui Gazem, Candance Eastep, Ahgean Davis, and Chuck Davies. We have read and agree to follow the COGS108 Team Policies.\n",
+"Our team members are Nicholas Campos, Lui Gazem, Cadence Eastep, Ahgean Davis, and Chuck Davies. We have read and agree to follow the COGS108 Team Policies.\n",
"\n",
"- **Clear Communication:** We will communicate primarily through group chat and scheduled meetings. All members are expected to respond within 24 hours unless previously communicated otherwise. Important decisions will be made during meetings to ensure transparency.\n",
"\n",
diff --git a/01-DataCheckpoint.ipynb b/01-DataCheckpoint.ipynb
index f2ef5ba..6bfc73f 100644
--- a/01-DataCheckpoint.ipynb
+++ b/01-DataCheckpoint.ipynb
@@ -3,26 +3,7 @@
{
"cell_type": "markdown",
"metadata": {},
- "source": [
- "## Rubric\n",
- "\n",
- "Instructions: DELETE this cell before you submit via a `git push` to your repo before deadline. This cell is for your reference only and is not needed in your report. \n",
- "\n",
- "Scoring: Out of 10 points\n",
- "\n",
- "- Each Developing => -2 pts\n",
- "- Each Unsatisfactory/Missing => -4 pts\n",
- " - until the score is \n",
- "\n",
- "If students address the detailed feedback in a future checkpoint they will earn these points back\n",
- "\n",
- "\n",
- "| | Unsatisfactory | Developing | Proficient | Excellent |\n",
- "|------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------|\n",
- "| Data relevance | Did not have data relevant to their question. Or the datasets don't work together because there is no way to line them up against each other. If there are multiple datasets, most of them have this trouble | Data was only tangentially relevant to the question or a bad proxy for the question. If there are multiple datasets, some of them may be irrelevant or can't be easily combined. | All data sources are relevant to the question. | Multiple data sources for each aspect of the project. It's clear how the data supports the needs of the project. |\n",
- "| Data description | Dataset or its cleaning procedures are not described. If there are multiple datasets, most have this trouble | Data was not fully described. If there are multiple datasets, some of them are not fully described | Data was fully described | The details of the data descriptions and perhaps some very basic EDA also make it clear how the data supports the needs of the project. |\n",
- "| Data wrangling | Did not obtain data. They did not clean/tidy the data they obtained. If there are multiple datasets, most have this trouble | Data was partially cleaned or tidied. Perhaps you struggled to verify that the data was clean because they did not present it well. If there are multiple datasets, some have this trouble | The data is cleaned and tidied. | The data is spotless and they used tools to visualize the data cleanliness and you were convinced at first glance |\n"
- ]
+ "source": []
},
{
"cell_type": "markdown",
@@ -37,23 +18,25 @@
"source": [
"## Authors\n",
"\n",
- "- Nicholas Campos: Conceptualization, Methodology, Writing - original draft\n",
- "- Lui Gazem: Analysis, Writing - original draft\n",
- "- Cadence Eastep: Background research, Writing - original draft, Writing - review & editing\n",
+ "- Nicholas Campos: Conceptualization, Methodology, Writing – original draft\n",
+ "\n",
+ "- Lui Gazem: Analysis, Writing – original draft\n",
+ "\n",
+ "- Cadence Eastep: Background research, Writing – original draft, Writing – review & editing\n",
+ "\n",
"- Ahgean Davis: Data curation, Software, Visualization\n",
- "- Chuck Davies: Project administration, Ethics review, Writing - review & editing"
+ "\n",
+ "- Chuck Davies: Project administration, Ethics review, Writing – review & editing\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Research Question\n",
- "\n",
- "**Primary research question:**\n",
+ "**Primary research question:** \n",
"How do daily COVID-19 case counts vary by day of the week at the county level, and how are these variations associated with mobility patterns (workplace, residential, retail & recreation, and transit mobility)?\n",
"\n",
- "**Secondary (backup) research question:**\n",
+ "**Secondary (backup) research question:** \n",
"To what extent are changes in workplace and residential mobility associated with subsequent changes in daily COVID-19 case counts over time at the county level?\n",
"\n",
"**Variables and controls:**\n",
@@ -64,6 +47,12 @@
"**Note:** We discussed our primary research question with the TA, who indicated it was appropriate but recommended including a backup question in case broader scope is needed after exploratory data analysis. Both questions use the same county-level COVID-19 and mobility dataset."
]
},
+ {
+ "attachments": {},
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -79,7 +68,7 @@
"\n",
"For our project, we aim to more closely examine variation in COVID-19 case counts across days of the week, with particular attention to identifying when case spikes are most pronounced. Consistent with our hypothesis, we anticipate that reported cases will be higher during weekdays compared to weekends. One possible explanation is that essential workers continued in-person employment throughout much of the pandemic, increasing opportunities for exposure during the workweek. Workplace environments may facilitate transmission due to repeated close interactions with coworkers and contact with shared surfaces (e.g., counters, equipment, communal appliances). These routine occupational exposures could contribute to elevated transmission risk during weekdays, which may subsequently be reflected in higher reported case counts. This hypothesis is supported by prior empirical work conducted by Hannah Hoffman and The COVID Tracking Project, which examined temporal reporting patterns in state-level COVID-19 case data. Their analysis demonstrated a consistent decline in reported case counts over weekends. Importantly, this pattern was attributed not only to behavioral changes but also to structural factors, including reduced testing, reporting delays, and public health regulations that limited activity during weekends in many states at the time.\n",
"\n",
- "In a study conducted by Bulut Boru and M. Emre Gursoy, the authors developed a computational model designed to identify and learn patterns between human mobility and subsequent COVID-19 case counts. Their approach dynamically selects the optimal time frame and predictive method to enhance accuracy. When evaluated across 13 countries, the model demonstrated a high degree of precision in forecasting daily COVID-19 cases. The study further revealed a strong correlation between the movement patterns of individuals and the spread of the virus within their communities. Findings of this nature underscore the substantial impact that human mobility can have on disease transmission, providing empirical justification for the strict travel restrictions and mobility-limiting policies implemented by governments worldwide during the pandemic."
+ "In a study conducted by Bulut Boru and M. Emre Gursoy, the authors developed a computational model designed to identify and learn patterns between human mobility and subsequent COVID-19 case counts. Their approach dynamically selects the optimal time frame and predictive method to enhance accuracy. When evaluated across 13 countries, the model demonstrated a high degree of precision in forecasting daily COVID-19 cases. The study further revealed a strong correlation between the movement patterns of individuals and the spread of the virus within their communities. Findings of this nature underscore the substantial impact that human mobility can have on disease transmission, providing empirical justification for the strict travel restrictions and mobility-limiting policies implemented by governments worldwide during the pandemic.\n"
]
},
{
@@ -93,7 +82,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "We hypothesize that daily COVID-19 case counts (daily_cases) will vary systematically across days of the week, with lower reported cases on weekends (is_weekend = 1) and higher reported cases during weekdays. Additionally, we predict that increases in workplace mobility (workplaces) will be positively associated with daily COVID-19 case counts, while increases in residential mobility (residential) will be negatively associated with daily case counts at the county level over time."
+ "We hypothesize that daily COVID-19 case counts (daily_cases) will vary systematically across days of the week, with lower reported cases on weekends (is_weekend = 1) and higher reported cases during weekdays. Additionally, we predict that increases in workplace mobility (workplaces) will be positively associated with daily COVID-19 case counts, while increases in residential mobility (residential) will be negatively associated with daily case counts at the county level over time.\n"
]
},
{
@@ -107,24 +96,32 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "### Data overview\n",
+ "### Data Overview\n",
"\n",
- "**Dataset #1: Integrated COVID-19 Case and Mobility Data**\n",
- "- **Dataset Name:** IntegratedData3 (county-level COVID-19 cases with Google Mobility data)\n",
- "- **Link to dataset:** [add link here]\n",
- "- **Number of observations:** approximately 1.2 million rows (one per county per day)\n",
- "- **Number of variables:** 16 columns\n",
- "- **Key variables:** date, county, state, fips, cases, deaths, daily_cases, daily_deaths, day_of_week, is_weekend, is_holiday, retail_recreation, grocery_pharmacy, parks, transit, workplaces, residential\n",
- "- **Shortcomings:** testing access varied across counties and time; mobility columns have substantial missingness in rural counties; negative daily_cases values appear due to data corrections\n",
+ "- **Dataset Name:** IntegratedData3 (County-Level COVID-19 Case and Mobility Dataset)\n",
+ "- **Link:** https://drive.google.com/uc?export=download&id=1hQwLJdeMyXGf2Fp0xEBXQrcj8sZsRYca\n",
+ "- **Number of observations:** 935,443 rows\n",
+ "- **Number of variables:** 17 columns\n",
+ "- **Relevant variables:** `daily_cases` (new COVID-19 cases per county per day — primary DV), `date` (YYYY-MM-DD), `fips` (unique county code used as merge key), `day_of_week` (0=Monday through 6=Sunday), `is_weekend` (binary flag), and mobility columns `workplaces`, `residential`, `retail_recreation`, `grocery_pharmacy`, `parks`, and `transit` (all percent change from pre-pandemic baseline).\n",
+ "- **Shortcomings:** Case counts depend on testing availability and weekend reporting drops artificially. Mobility columns `parks` and `transit` have high missingness in rural counties. Mobility data underrepresents people without smartphones.\n",
"\n",
- "This is a single pre-merged dataset combining COVID-19 case counts with Google Community Mobility Reports. The join was performed on date and fips (county FIPS code), producing one row per county per day."
+ "This project uses a single pre-integrated dataset where COVID-19 case data and Google mobility data were joined using the `fips` county code as the primary key. Because the data arrived pre-merged, we focus our wrangling on cleaning and validating the integrated file rather than performing the join ourselves."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 39,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The autoreload extension is already loaded. To reload it, use:\n",
+ " %reload_ext autoreload\n"
+ ]
+ }
+ ],
"source": [
"# Run this code every time when you're actively developing modules in .py files.\n",
"%load_ext autoreload\n",
@@ -133,9 +130,48 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 53,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Overall Download Progress: 0%| | 0/1 [00:00\n",
+ "\n",
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "**Cleaning summary:** The raw dataset contained one row per US county per day. The `transit` and `parks` columns had the highest missingness, confirmed to be systematic rather than random — counties with missing mobility data tend to be smaller/less-populated where Google had insufficient location data. We retained these NaN values rather than imputing, since filling them would misrepresent a true absence of data. A small number of rows had negative `daily_cases` values resulting from retroactive corrections by county health departments — these were flagged with a `negative_case_flag` column and a new `daily_cases_clean` column was created with negatives clipped to zero for analysis. No duplicate county-date rows were found. The dataset is tidy. The cleaned file was saved to `data/02-processed/covid_mobility_clean.csv`."
+ "# EDA Plot 1: Distribution of daily_cases_clean\n",
+ "plt.figure(figsize=(8, 4))\n",
+ "df_clean[df_clean['daily_cases_clean'] < 500]['daily_cases_clean'].hist(bins=50, color='steelblue', edgecolor='black')\n",
+ "plt.xlabel('Daily Cases (filtered below 500 for readability)')\n",
+ "plt.ylabel('Frequency')\n",
+ "plt.title('Distribution of Daily COVID-19 Cases (County-Day Level)')\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n",
+ "\n",
+ "# EDA Plot 2: Average daily cases by day of week\n",
+ "dow_avg = df_clean.groupby('day_of_week')['daily_cases_clean'].mean()\n",
+ "days = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']\n",
+ "plt.figure(figsize=(7, 4))\n",
+ "plt.bar(days, dow_avg.values, color='coral', edgecolor='black')\n",
+ "plt.xlabel('Day of Week')\n",
+ "plt.ylabel('Average Daily Cases')\n",
+ "plt.title('Average Daily COVID-19 Cases by Day of Week')\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "### Dataset #2\n",
- "\n",
- "See instructions above for Dataset #1. Feel free to keep adding as many more datasets as you need. Put each new dataset in its own section just like these.\n",
- "\n",
- "Lastly if you do have multiple datasets, add another section where you demonstrate how you will join, align, cross-reference or whatever to combine data from the different datasets.\n",
- "\n",
- "Please note that you can always keep adding more datasets in the future if these datasets you turn in for the checkpoint aren't sufficient. The goal here is demonstrate that you can obtain and wrangle data. You are not tied down to only use what you turn in right now."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
- "source": [
- "## YOUR CODE TO LOAD/CLEAN/TIDY/WRANGLE THE DATA GOES HERE\n"
+ "**Cleaning summary:** The raw dataset contained one row per US county per day. The `transit` and `parks` columns had the highest missingness, confirmed to be systematic - counties with missing mobility data tend to be smaller where Google had insufficient location data. We retained NaN values rather than imputing since filling them would misrepresent a true absence of data. Negative `daily_cases` values from retroactive corrections were flagged and clipped to zero in `daily_cases_clean`. No duplicate county-date rows were found. The cleaned file was saved to `data/02-processed/covid_mobility_clean.csv`.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Ethics\n",
+ "## Ethics \n",
+ "\n",
+ "Instructions: Keep the contents of this cell. For each item on the checklist\n",
+ "- put an X there if you've considered the item\n",
+ "- IF THE ITEM IS RELEVANT place a short paragraph after the checklist item discussing the issue.\n",
+ " \n",
+ "Items on this checklist are meant to provoke discussion among good-faith actors who take their ethical responsibilities seriously. Your teams will document these discussions and decisions for posterity using this section. You don't have to solve these problems, you just have to acknowledge any potential harm no matter how unlikely.\n",
+ "\n",
+ "Here is a [list of real world examples](https://deon.drivendata.org/examples/) for each item in the checklist that can refer to.\n",
+ "\n",
+ "[](http://deon.drivendata.org/)\n",
"\n",
"### A. Data Collection\n",
- " - [X] **A.1 Informed consent**\n",
+ " - [X] **A.1 Informed consent**: If there are human subjects, have they given informed consent, where subjects affirmatively opt-in and have a clear understanding of the data uses to which they consent?\n",
"\n",
"Our analysis uses publicly available, aggregated county-level COVID-19 case counts and mobility metrics. We do not recruit participants, interact with individuals, or collect new individual-level data, so informed consent is not required for our work. However, we acknowledge that mobility metrics originate from smartphone location signals and that many users may not fully understand downstream secondary uses, even when data is anonymized and aggregated.\n",
"\n",
- " - [X] **A.2 Collection bias**\n",
+ " - [X] **A.2 Collection bias**: Have we considered sources of bias that could be introduced during data collection and survey design and taken steps to mitigate those?\n",
"\n",
"Mobility data can underrepresent people without smartphones or stable location services (for example older adults, low income communities, unhoused populations, and some rural areas). COVID-19 case counts also reflect testing access and reporting practices that vary by county and over time. We will interpret findings as population-level associations and avoid claims that assume equal measurement quality across all counties or groups.\n",
"\n",
- " - [X] **A.3 Limit PII exposure**\n",
+ " - [X] **A.3 Limit PII exposure**: Have we considered ways to minimize exposure of personally identifiable information (PII) for example through anonymization or not collecting information that isn't relevant for analysis?\n",
"\n",
- "Our dataset is already aggregated to county-day level and does not include direct identifiers (names, addresses, device IDs). We will not attempt any re-identification or linkage to external individual-level datasets.\n",
+ "Our dataset is already aggregated to county-day level and does not include direct identifiers (names, addresses, device IDs). We will not attempt any re-identification or linkage to external individual-level datasets. We will also avoid presenting results as county rankings intended to label places as good or bad.\n",
"\n",
- " - [X] **A.4 Downstream bias mitigation**\n",
+ " - [X] **A.4 Downstream bias mitigation**: Have we considered ways to enable testing downstream results for biased outcomes (e.g., collecting data on protected group status like race or gender)?\n",
"\n",
- "Our results could be misused to blame specific regions or communities for COVID-19 spread. We will present conclusions carefully, emphasizing that mobility is a proxy for many structural factors and that associations do not imply individual fault or causation.\n",
+ "Our results could be misused to blame specific regions or communities for COVID-19 spread or to justify punitive policy based on mobility patterns. We will present conclusions carefully, emphasizing that mobility is a proxy for many structural factors and that associations do not imply individual fault or causation. Where feasible, we will add contextual covariates (for example population, time controls, fixed effects) and include limitations that discourage stigmatizing interpretations.\n",
"\n",
"### B. Data Storage\n",
- " - [X] **B.1 Data security**\n",
+ " - [X] **B.1 Data security**: Do we have a plan to protect and secure data (e.g., encryption at rest and in transit, access controls on internal users and third parties, access logs, and up-to-date software)?\n",
"\n",
- "The data are public and aggregated. We will keep data in the course GitHub repo and will not store any private keys or credentials in the repository.\n",
+ "The data are public and aggregated, so the risk level is low, but we will still follow basic security practices: keep data in the course GitHub repo, avoid sharing raw data outside the team, and ensure devices/accounts used to access the repo are protected. We will not store any private keys or credentials in the repository.\n",
"\n",
- " - [X] **B.2 Right to be forgotten**\n",
+ " - [X] **B.2 Right to be forgotten**: Do we have a mechanism through which an individual can request their personal information be removed?\n",
"\n",
- "Because the dataset does not contain individual-level identifiers, individuals cannot be singled out or removed from our data.\n",
+ "Because the dataset does not contain individual-level identifiers, individuals cannot be singled out or removed from our data. We rely on the data providers' governance and opt-out mechanisms (if any). Our project does not maintain user-level records.\n",
"\n",
- " - [X] **B.3 Data retention plan**\n",
+ " - [X] **B.3 Data retention plan**: Is there a schedule or plan to delete the data after it is no longer needed?\n",
"\n",
- "We will keep the dataset only as long as needed for the course and reproducibility of the final report.\n",
+ "We will keep the dataset only as long as needed for the course and reproducibility of the final report. After the course, we can remove large raw data files from the repo while retaining derived results and code that do not contain sensitive information.\n",
"\n",
"### C. Analysis\n",
- " - [X] **C.1 Missing perspectives**\n",
+ " - [X] **C.1 Missing perspectives**: Have we sought to address blindspots in the analysis through engagement with relevant stakeholders (e.g., checking assumptions and discussing implications with affected communities and subject matter experts)?\n",
"\n",
- "Public health outcomes depend on factors not captured by mobility (healthcare access, workplace protections, housing density, local policy, and reporting practices). We will check our assumptions using public health context from reputable sources.\n",
+ "Public health outcomes depend on factors not captured by mobility (healthcare access, workplace protections, housing density, local policy, and reporting practices). We will check our assumptions using public health context from reputable sources and discuss limitations so our results are not interpreted as purely behavioral explanations.\n",
"\n",
- " - [X] **C.2 Dataset bias**\n",
+ " - [X] **C.2 Dataset bias**: Have we examined the data for possible sources of bias and taken steps to mitigate or address these biases (e.g., stereotype perpetuation, confirmation bias, imbalanced classes, or omitted confounding variables)?\n",
"\n",
- "We expect bias from uneven testing/reporting, time-varying policy changes, and demographic differences across counties. We will use appropriate controls (time effects, county fixed effects, lagged outcomes where relevant).\n",
+ "We expect bias from uneven testing/reporting, time-varying policy changes, and demographic differences across counties. We will use appropriate controls (time effects, county fixed effects, lagged outcomes where relevant) and perform basic robustness checks (for example examining patterns across time periods) to reduce the chance of misleading conclusions.\n",
"\n",
- " - [X] **C.3 Honest representation**\n",
+ " - [X] **C.3 Honest representation**: Are our visualizations, summary statistics, and reports designed to honestly represent the underlying data?\n",
"\n",
- "We will avoid causal wording and report results as associations. Visualizations will clearly label units and transformations.\n",
+ "We will avoid causal wording and report results as associations. Visualizations will clearly label units and transformations, and we will include uncertainty or variability where possible. We will not cherry-pick dates or counties to support a preferred conclusion.\n",
"\n",
- " - [X] **C.4 Privacy in analysis**\n",
+ " - [X] **C.4 Privacy in analysis**: Have we ensured that data with PII are not used or displayed unless necessary for the analysis?\n",
"\n",
- "No PII is present in the dataset and none will be introduced.\n",
+ "No PII is present in the dataset and none will be introduced. We will only analyze and display aggregated county-level metrics.\n",
"\n",
- " - [X] **C.5 Auditability**\n",
+ " - [X] **C.5 Auditability**: Is the process of generating the analysis well documented and reproducible if we discover issues in the future?\n",
"\n",
- "All preprocessing, analysis steps, and figure generation will be documented in notebooks/scripts in the repository.\n",
+ "All preprocessing, analysis steps, and figure generation will be documented in notebooks/scripts in the repository. This supports reproducibility and makes it easier to identify issues if errors or data problems are later discovered.\n",
"\n",
"### D. Modeling\n",
- " - [X] **D.1 Proxy discrimination**\n",
+ " - [X] **D.1 Proxy discrimination**: Have we ensured that the model does not rely on variables or proxies for variables that are unfairly discriminatory?\n",
"\n",
- "Mobility measures can correlate with socioeconomic status and occupation types. We will not use models for individual-level prediction.\n",
+ "Mobility measures can correlate with socioeconomic status and occupation types. We will not use models for individual-level prediction, and we will interpret results as structural and measurement patterns rather than traits of any group.\n",
"\n",
- " - [X] **D.2 Fairness across groups**\n",
+ " - [X] **D.2 Fairness across groups**: Have we tested model results for fairness with respect to different affected groups (e.g., tested for disparate error rates)?\n",
"\n",
- "We are not building a high-stakes decision system, but we will check whether model fit varies across counties and report differences as limitations.\n",
+ "We are not building a high-stakes decision system, but we can still check whether model residuals or fit vary across counties (for example by population size or region). If large differences appear, we will report them as limitations and avoid overgeneralizing.\n",
"\n",
- " - [X] **D.3 Metric selection**\n",
+ " - [X] **D.3 Metric selection**: Have we considered the effects of optimizing for our defined metrics and considered additional metrics?\n",
"\n",
- "We will use standard regression evaluation but will prioritize interpretability over optimizing a single metric.\n",
+ "We will use standard regression evaluation (for example goodness of fit and error measures) but will prioritize interpretability and stability over optimizing a single metric. We will compare multiple reasonable specifications to ensure conclusions are not driven by one modeling choice.\n",
"\n",
- " - [X] **D.4 Explainability**\n",
+ " - [X] **D.4 Explainability**: Can we explain in understandable terms a decision the model made in cases where a justification is needed?\n",
"\n",
- "We will use interpretable models (linear regression with coefficients) and explain findings in plain language.\n",
+ "We will use interpretable models (for example linear regression with coefficients) and explain findings in plain language. We will tie interpretations directly to observed variables and note when the data do not justify a stronger claim.\n",
"\n",
- " - [X] **D.5 Communicate limitations**\n",
+ " - [X] **D.5 Communicate limitations**: Have we communicated the shortcomings, limitations, and biases of the model to relevant stakeholders in ways that can be generally understood?\n",
"\n",
- "Our report will include a limitations section emphasizing that associations are not causation and mobility metrics may not represent all populations equally.\n",
+ "Our report will include a limitations section emphasizing that associations are not causation, case reporting varies by testing/reporting practices, and mobility metrics may not represent all populations equally. We will explicitly discourage stigmatizing or punitive interpretations.\n",
"\n",
"### E. Deployment\n",
- " - [X] **E.1 Monitoring and evaluation**\n",
+ " - [X] **E.1 Monitoring and evaluation**: Do we have a clear plan to monitor the model and its impacts after it is deployed (e.g., performance monitoring, regular audit of sample predictions, human review of high-stakes decisions, reviewing downstream impacts of errors or low-confidence decisions, testing for concept drift)?\n",
"\n",
- "We are not deploying a model in production. We will sanity-check outputs and document cleaning decisions.\n",
+ "We are not deploying a model in production. Still, we will sanity-check outputs (for example extreme spikes, missing dates, or inconsistent values) and document cleaning decisions to reduce errors in the final report.\n",
"\n",
- " - [X] **E.2 Redress**\n",
+ " - [X] **E.2 Redress**: Have we discussed with our organization a plan for response if users are harmed by the results (e.g., how does the data science team evaluate these cases and update analysis and models to prevent future harm)?\n",
"\n",
- "This is a course research analysis, not a deployed system. The main risk is misinterpretation, which we mitigate through careful framing and clearly stating limitations.\n",
+ "Because this is a course research analysis and not a deployed system, formal redress is not applicable. The main risk is misinterpretation, so we mitigate harm through careful framing, avoiding causal claims, and clearly stating limitations and appropriate uses.\n",
"\n",
- " - [X] **E.3 Roll back**\n",
+ " - [X] **E.3 Roll back**: Is there a way to turn off or roll back the model in production if necessary?\n",
"\n",
- "Not applicable. If issues are found, we can update or remove analyses in the repository.\n",
+ "Not applicable because we are not deploying a production model. If issues are found, we can update or remove analyses in the repository and clearly note changes.\n",
"\n",
- " - [X] **E.4 Unintended use**\n",
+ " - [X] **E.4 Unintended use**: Have we taken steps to identify and prevent unintended uses and abuse of the model and do we have a plan to monitor these once the model is deployed?\n",
"\n",
- "Unintended use is possible if results are used to justify surveillance or punitive restrictions. We will avoid producing county rankings meant for enforcement decisions."
+ "Unintended use is possible if results are used to justify surveillance or punitive restrictions targeted at particular counties. We will avoid producing county rankings meant for enforcement decisions, use neutral language, and emphasize that mobility is an imperfect proxy that should not be used for punitive action.\n"
]
},
{
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
+ {
+ "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
@@ -477,29 +1207,72 @@
"\n",
"Our team members are Nicholas Campos, Lui Gazem, Cadence Eastep, Ahgean Davis, and Chuck Davies. We have read and agree to follow the COGS108 Team Policies.\n",
"\n",
- "- **Clear Communication:** We will communicate primarily through group chat and scheduled meetings. All members are expected to respond within 24 hours unless previously communicated otherwise.\n",
- "- **Accountability & Deadlines:** Each member is responsible for completing their assigned tasks on time.\n",
- "- **Equal Contribution:** We expect all members to contribute meaningfully to research, coding, writing, and revisions.\n",
- "- **Respectful Conflict Resolution:** If disagreements arise, we will address them respectfully and directly.\n",
- "- **Quality Control:** Before submission, all members will review the project to ensure clarity, correctness, and completeness."
+ "- **Clear Communication:** We will communicate primarily through group chat and scheduled meetings. All members are expected to respond within 24 hours unless previously communicated otherwise. Important decisions will be made during meetings to ensure transparency.\n",
+ "\n",
+ "- **Accountability & Deadlines:** Each member is responsible for completing their assigned tasks on time. If someone anticipates difficulty meeting a deadline, they will notify the group as early as possible so adjustments can be made.\n",
+ "\n",
+ "- **Equal Contribution:** We expect all members to contribute meaningfully to research, coding, writing, and revisions. Work will be divided fairly, but we will support one another if someone needs assistance.\n",
+ "\n",
+ "- **Respectful Conflict Resolution:** If disagreements arise, we will address them respectfully and directly. We will focus on the work rather than personal criticism and aim to resolve issues collaboratively.\n",
+ "\n",
+ "- **Quality Control:** Before submission, all members will review the project to ensure clarity, correctness, and completeness. No section will be submitted without at least one additional team member reviewing it.\n",
+ "\n",
+ "We are committed to maintaining professionalism, consistency, and mutual respect throughout the project.\n"
]
},
{
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
+ {
+ "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## Project Timeline Proposal\n",
"\n",
- "Our team is currently in Week 7 of the quarter (beginning February 15). The final written report and presentation video are due in Week 10. We meet every Sunday at 2 PM to review progress and assign next steps.\n",
+ "Our team is currently in Week 7 of the quarter (beginning February 15). The final written report and presentation video are due in Week 10. We meet every Sunday at 2 PM to review progress and assign next steps. Our remaining timeline is as follows:\n",
"\n",
"| Meeting Date | Meeting Time | Completed Before Meeting | Discuss at Meeting |\n",
- "|---|---|---|---|\n",
+ "|--------------|--------------|--------------------------|--------------------|\n",
"| February 16 | 2 PM | Finalize data cleaning; Complete full exploratory data analysis | Confirm final analysis plan; Assign statistical testing and visualization tasks |\n",
"| February 23 | 2 PM | Complete statistical analysis; Generate all visualizations | Interpret results; Identify key findings; Begin drafting Results section |\n",
"| March 2 | 2 PM | Draft Introduction, Methods, and Results sections | Peer review draft; Revise figures; Outline presentation video structure |\n",
"| March 9 | 2 PM | Draft Discussion and Conclusion; Create presentation slides | Final content edits; Record presentation video |\n",
- "| Week 10 Deadline | Before submission time | Final formatting and proofreading | Submit final written report and presentation video |"
+ "| Week 10 Deadline | Before submission time | Final formatting and proofreading | Submit final written report and presentation video |\n",
+ "\n",
+ "Responsibilities for analysis, writing, visualization, and video production will be divided evenly among team members. Each major section will be reviewed by at least one additional member prior to submission to ensure clarity and accuracy.\n",
+ "\n",
+ "We do not anticipate requiring tools beyond those covered in COGS 108 (Python, pandas, visualization libraries, and standard statistical methods). If additional techniques become necessary, we will consult course materials and TA guidance before implementation.\n"
]
+ },
+ {
+ "attachments": {},
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
}
],
"metadata": {
@@ -518,7 +1291,12 @@
"name": "python",
"nbformat_minor": 4,
"pygments_lexer": "ipython3",
- "version": "3.11.13"
+ "version": "3.11.9"
+ },
+ "vscode": {
+ "interpreter": {
+ "hash": "16b860a9f5fc21240e9d88c0ee13691518c3ce67be252e54a03b9b5b11bd3c7a"
+ }
}
},
"nbformat": 4,
diff --git a/02-EDACheckpoint.ipynb b/02-EDACheckpoint.ipynb
index 0f5a34f..23fcfb4 100644
--- a/02-EDACheckpoint.ipynb
+++ b/02-EDACheckpoint.ipynb
@@ -1,41 +1,58 @@
{
"cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Rubric\n",
- "\n",
- "Instructions: DELETE this cell before you submit via a `git push` to your repo before deadline. This cell is for your reference only and is not needed in your report. \n",
- "\n",
- " Scoring: Out of 10 points\n",
- "\n",
- "- Each Developing => -2 pts\n",
- "- Each Unsatisfactory/Missing => -4 pts\n",
- " - until the score is 0\n",
- "\n",
- "If students address the detailed feedback in a future checkpoint they will earn these points back\n",
+ "# COGS 108 - EDA Checkpoint"
+ ]
+ },
+ {
+ "attachments": {},
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Authors\n",
"\n",
+ "- Nicholas Campos: Conceptualization, Methodology, Writing – original draft\n",
"\n",
+ "- Lui Gazem: Analysis, Writing – original draft\n",
"\n",
+ "- Cadence Eastep: Background research, Writing – original draft, Writing – review & editing\n",
"\n",
+ "- Ahgean Davis: Data curation, Software, Visualization\n",
"\n",
+ "- Chuck Davies: Project administration, Ethics review, Writing – review & editing\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+"**Primary research question:** \n",
+ "How do daily COVID-19 case counts vary by day of the week at the county level, and how are these variations associated with mobility patterns (workplace, residential, retail & recreation, and transit mobility)?\n",
"\n",
- "| | **Unsatisfactory** | **Developing** | **Proficient** | **Excellent** |\n",
- "|----------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n",
- "| **EDA relevance** | EDA is mostly neither relevant to the question nor helpful in figuring out how to address the question. Or the EDA does address the question, but many obviously relevant variables / analyses / figures were not included. | EDA is partly irrelevant/unhelpful. EDA missed one or two obvioulsy relevant analysis (distributions of single variables or relationships between variables) | EDA includes the obviously relevant / helpful variables in addressing the question. | Thorough EDA fully explored the dataset |\n",
- "| **EDA analysis and description** | Many of the analyses are poor choices (e.g., using means instead of medians for obviously skewed data), or are poorly described in the text, or do not aid understanding the data | Some of the analyses are poor choices, or are poorly described in the text, or do not aid understanding the data | All analyses are correct choices. Only one or two have minor issues in the text descriptions supporting them. Mostly they fit well with other elements of the EDA and support understanding the data | All analyses are correct choices with clear text descriptions supporting them. The figures fit well with the other elements of the EDA, producing a clear understanding of the data. |\n",
- "| **EDA figures** | Many of the figures are poor plot choices (e.g., using a bar plot to represent a time series where it would be better to use a line plot) or have poor aesthetics (including colormap, data point shape/color, axis labels, titles, annotations, text legibility) or do not aid understanding the data | Some of the figures are poor plot choices or have poor aesthetics. Some figures do not aid understanding the data | All figures are correct plot choices. Only one or two have minor questionable aesthetic choices. The figures mostly fit well with the other elements of the EDA and support understanding the data | All figures are correct plot choices with beautiful aesthetics. The figures fit well with the other elements of the EDA, producing a clear understanding of the data. |\n",
+ "**Secondary (backup) research question:** \n",
+ "To what extent are changes in workplace and residential mobility associated with subsequent changes in daily COVID-19 case counts over time at the county level?\n",
"\n",
+ "**Variables and controls:**\n",
+ "- **Dependent variable (DV):** daily new COVID-19 case counts (county-day level)\n",
+ "- **Independent variables (IVs):** workplace mobility, residential mobility, retail & recreation mobility, transit mobility, day of week\n",
+ "- **Controls:** county fixed effects, time (date or week index), population (if available), lagged case counts\n",
"\n",
- "\n"
+ "**Note:** We discussed our primary research question with the TA, who indicated it was appropriate but recommended including a backup question in case broader scope is needed after exploratory data analysis. Both questions use the same county-level COVID-19 and mobility dataset."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "# COGS 108 - EDA Checkpoint"
+ "## Background and Prior Work"
]
},
{
@@ -43,25 +60,24 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Authors\n",
+ "Mobility rates and COVID-19 case counts exhibited a substantial correlation throughout the pandemic. In response to the rapid spread of the virus, public health authorities urged individuals to remain at home whenever possible, recognizing that increased movement and social interaction heightened the risk of transmission. From the onset of this global crisis, policymakers closely monitored the relationship between population mobility and infection rates in an effort to mitigate uncontrolled surges in cases. A study conducted by the Centers for Disease Control and Prevention (CDC) examined anonymized cell phone GPS data to assess patterns of movement across counties in the United States. The researchers observed that reductions in trips outside the home were associated with subsequent declines in reported COVID-19 cases. Specifically, changes in mobility were analyzed in relation to new case counts approximately 11 days later, reflecting the typical incubation period and reporting delays. The study also compared urban and rural areas to determine whether mobility patterns affected transmission differently across settings. This distinction was critical, as urban centers tend to be more densely populated, exhibit distinct transportation behaviors, and may experience differing timelines of viral introduction and spread compared to rural communities.\n",
"\n",
- "Instructions: REPLACE the contents of this cell with your team list and their contributions. Note that this will change over the course of the checkpoints\n",
+ "For our project, we aim to more closely examine variation in COVID-19 case counts across days of the week, with particular attention to identifying when case spikes are most pronounced. Consistent with our hypothesis, we anticipate that reported cases will be higher during weekdays compared to weekends. One possible explanation is that essential workers continued in-person employment throughout much of the pandemic, increasing opportunities for exposure during the workweek. Workplace environments may facilitate transmission due to repeated close interactions with coworkers and contact with shared surfaces (e.g., counters, equipment, communal appliances). These routine occupational exposures could contribute to elevated transmission risk during weekdays, which may subsequently be reflected in higher reported case counts. This hypothesis is supported by prior empirical work conducted by Hannah Hoffman and The COVID Tracking Project, which examined temporal reporting patterns in state-level COVID-19 case data. Their analysis demonstrated a consistent decline in reported case counts over weekends. Importantly, this pattern was attributed not only to behavioral changes but also to structural factors, including reduced testing, reporting delays, and public health regulations that limited activity during weekends in many states at the time.\n",
"\n",
- "This is a modified [CRediT taxonomy of contributions](https://credit.niso.org). For each group member please list how they contributed to this project using these terms:\n",
- "> Analysis, Background research, Conceptualization, Data curation, Experimental investigation, Methodology, Project administration, Software, Visualization, Writing – original draft, Writing – review & editing\n",
- "\n",
- "Example team list and credits:\n",
- "- Alice Anderson: Conceptualization, Data curation, Methodology, Writing - original draft\n",
- "- Bob Barker: Analysis, Software, Visualization\n",
- "- Charlie Chang: Project administration, Software, Writing - review & editing\n",
- "- Dani Delgado: Analysis, Background research, Visualization, Writing - original draft"
+ "In a study conducted by Bulut Boru and M. Emre Gursoy, the authors developed a computational model designed to identify and learn patterns between human mobility and subsequent COVID-19 case counts. Their approach dynamically selects the optimal time frame and predictive method to enhance accuracy. When evaluated across 13 countries, the model demonstrated a high degree of precision in forecasting daily COVID-19 cases. The study further revealed a strong correlation between the movement patterns of individuals and the spread of the virus within their communities. Findings of this nature underscore the substantial impact that human mobility can have on disease transmission, providing empirical justification for the strict travel restrictions and mobility-limiting policies implemented by governments worldwide during the pandemic.\n"
]
},
+ {
+ "attachments": {},
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Research Question"
+ "## Hypothesis\n"
]
},
{
@@ -69,188 +85,1037 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "Instructions: REPLACE the contents of this cell with your work, including any updates to recover points lost in your proposal feedback\n",
- "\n"
+ "We hypothesize that daily COVID-19 case counts (daily_cases) will vary systematically across days of the week, with lower reported cases on weekends (is_weekend = 1) and higher reported cases during weekdays. Additionally, we predict that increases in workplace mobility (workplaces) will be positively associated with daily COVID-19 case counts, while increases in residential mobility (residential) will be negatively associated with daily case counts at the county level over time.\n"
]
},
{
+ "attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Background and Prior Work"
]
},
{
"attachments": {},
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": []
+ },
+ {
"cell_type": "markdown",
"metadata": {},
"source": [
- "Instructions: REPLACE the contents of this cell with your work, including any updates to recover points lost in your proposal feedback"
+ "## Data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Hypothesis\n"
+ "### Data Overview\n",
+ "\n",
+ "- **Dataset Name:** IntegratedData3 (County-Level COVID-19 Case and Mobility Dataset)\n",
+ "- **Link:** https://drive.google.com/uc?export=download&id=1hQwLJdeMyXGf2Fp0xEBXQrcj8sZsRYca\n",
+ "- **Number of observations:** 935,443 rows\n",
+ "- **Number of variables:** 17 columns\n",
+ "- **Relevant variables:** `daily_cases` (new COVID-19 cases per county per day — primary DV), `date` (YYYY-MM-DD), `fips` (unique county code used as merge key), `day_of_week` (0=Monday through 6=Sunday), `is_weekend` (binary flag), and mobility columns `workplaces`, `residential`, `retail_recreation`, `grocery_pharmacy`, `parks`, and `transit` (all percent change from pre-pandemic baseline).\n",
+ "- **Shortcomings:** Case counts depend on testing availability and weekend reporting drops artificially. Mobility columns `parks` and `transit` have high missingness in rural counties. Mobility data underrepresents people without smartphones.\n",
+ "\n",
+ "This project uses a single pre-integrated dataset where COVID-19 case data and Google mobility data were joined using the `fips` county code as the primary key. Because the data arrived pre-merged, we focus our wrangling on cleaning and validating the integrated file rather than performing the join ourselves."
]
},
{
- "attachments": {},
- "cell_type": "markdown",
+ "cell_type": "code",
+ "execution_count": 32,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The autoreload extension is already loaded. To reload it, use:\n",
+ " %reload_ext autoreload\n"
+ ]
+ }
+ ],
+ "source": [
+ "# Run this code every time when you're actively developing modules in .py files. It's not needed if you aren't making modules\n",
+ "#\n",
+ "## this code is necessary for making sure that any modules we load are updated here \n",
+ "## when their source code .py files are modified\n",
+ "\n",
+ "%load_ext autoreload\n",
+ "%autoreload 2"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 53,
"metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Overall Download Progress: 0%| | 0/1 [00:00\n",
+ "\n",
+ "
"
+ ],
+ "text/plain": [
+ " date county state fips cases deaths daily_cases \\\n",
+ "0 2021-01-01 autauga alabama 1001 4239.0 50.0 0.0 \n",
+ "1 2021-01-02 autauga alabama 1001 4268.0 50.0 29.0 \n",
+ "2 2021-01-03 autauga alabama 1001 4305.0 50.0 37.0 \n",
+ "3 2021-01-04 autauga alabama 1001 4336.0 50.0 31.0 \n",
+ "4 2021-01-05 autauga alabama 1001 4546.0 50.0 210.0 \n",
+ "\n",
+ " daily_deaths day_of_week is_weekend is_holiday retail_recreation \\\n",
+ "0 0.0 4.0 0.0 1.0 -46.0 \n",
+ "1 0.0 5.0 1.0 0.0 -10.0 \n",
+ "2 0.0 6.0 1.0 0.0 -12.0 \n",
+ "3 0.0 0.0 0.0 0.0 4.0 \n",
+ "4 0.0 1.0 0.0 0.0 2.0 \n",
+ "\n",
+ " grocery_pharmacy parks transit workplaces residential \\\n",
+ "0 -28.0 NaN NaN -77.0 28.0 \n",
+ "1 -5.0 NaN NaN -20.0 7.0 \n",
+ "2 -3.0 -10.0 NaN -14.0 6.0 \n",
+ "3 2.0 NaN NaN -26.0 8.0 \n",
+ "4 3.0 NaN NaN -22.0 8.0 \n",
+ "\n",
+ " negative_case_flag daily_cases_clean \n",
+ "0 0 0.0 \n",
+ "1 0 29.0 \n",
+ "2 0 37.0 \n",
+ "3 0 31.0 \n",
+ "4 0 210.0 "
+ ]
+ },
+ "execution_count": 59,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Clean the dataset\n",
+ "df_clean = df.copy()\n",
+ "df_clean = df_clean.drop_duplicates()\n",
+ "print('After dropping duplicates:', df_clean.shape[0], 'rows')\n",
+ "\n",
+ "# Flag negative daily_cases - keep rows but mark them\n",
+ "df_clean['negative_case_flag'] = (df_clean['daily_cases'] < 0).astype(int)\n",
+ "print('Rows flagged as negative daily_cases:', df_clean['negative_case_flag'].sum())\n",
"\n",
- "Instructions: REPLACE the contents of this cell with your work, including any updates to recover points lost in your data checkpoint feedback\n"
+ "# Clip negatives to 0 for analysis\n",
+ "df_clean['daily_cases_clean'] = df_clean['daily_cases'].clip(lower=0)\n",
+ "\n",
+ "# Drop temp columns\n",
+ "df_clean = df_clean.drop(columns=['transit_missing', 'parks_missing'])\n",
+ "print('Final clean dataset shape:', df_clean.shape)\n",
+ "df_clean.head()\n"
]
},
{
"cell_type": "code",
- "execution_count": 32,
+ "execution_count": 60,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "The autoreload extension is already loaded. To reload it, use:\n",
- " %reload_ext autoreload\n"
+ "Unique county-date pairs: 931458\n",
+ "Total rows: 935442\n",
+ "Warning: duplicate county-date rows exist.\n"
]
}
],
"source": [
- "# Run this code every time when you're actively developing modules in .py files. It's not needed if you aren't making modules\n",
- "#\n",
- "## this code is necessary for making sure that any modules we load are updated here \n",
- "## when their source code .py files are modified\n",
- "\n",
- "%load_ext autoreload\n",
- "%autoreload 2"
+ "# Verify tidy - one row per county per date\n",
+ "unique_pairs = df_clean.drop_duplicates(subset=['date', 'fips']).shape[0]\n",
+ "total_rows = df_clean.shape[0]\n",
+ "print('Unique county-date pairs:', unique_pairs)\n",
+ "print('Total rows:', total_rows)\n",
+ "if unique_pairs == total_rows:\n",
+ " print('Dataset is tidy - one row per county per date.')\n",
+ "else:\n",
+ " print('Warning: duplicate county-date rows exist.')\n"
]
},
{
"cell_type": "code",
- "execution_count": 33,
+ "execution_count": 61,
"metadata": {},
"outputs": [
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "Overall Download Progress: 0%| | 0/2 [00:00\n",
+ "\n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
\n",
+ "
daily_cases_clean
\n",
+ "
workplaces
\n",
+ "
residential
\n",
+ "
retail_recreation
\n",
+ "
transit
\n",
+ "
is_weekend
\n",
+ "
\n",
+ " \n",
+ " \n",
+ "
\n",
+ "
count
\n",
+ "
935442.00
\n",
+ "
924893.00
\n",
+ "
573705.00
\n",
+ "
591076.00
\n",
+ "
347338.00
\n",
+ "
935442.00
\n",
+ "
\n",
+ "
\n",
+ "
mean
\n",
+ "
35.55
\n",
+ "
-19.45
\n",
+ "
4.87
\n",
+ "
0.74
\n",
+ "
-1.77
\n",
+ "
0.24
\n",
+ "
\n",
+ "
\n",
+ "
std
\n",
+ "
238.71
\n",
+ "
13.63
\n",
+ "
4.52
\n",
+ "
19.62
\n",
+ "
33.76
\n",
+ "
0.43
\n",
+ "
\n",
+ "
\n",
+ "
min
\n",
+ "
0.00
\n",
+ "
-100.00
\n",
+ "
-29.00
\n",
+ "
-100.00
\n",
+ "
-100.00
\n",
+ "
0.00
\n",
+ "
\n",
+ "
\n",
+ "
25%
\n",
+ "
0.00
\n",
+ "
-25.00
\n",
+ "
2.00
\n",
+ "
-10.00
\n",
+ "
-25.00
\n",
+ "
0.00
\n",
+ "
\n",
+ "
\n",
+ "
50%
\n",
+ "
4.00
\n",
+ "
-18.00
\n",
+ "
4.00
\n",
+ "
0.00
\n",
+ "
-3.00
\n",
+ "
0.00
\n",
+ "
\n",
+ "
\n",
+ "
75%
\n",
+ "
20.00
\n",
+ "
-11.00
\n",
+ "
7.00
\n",
+ "
11.00
\n",
+ "
17.00
\n",
+ "
0.00
\n",
+ "
\n",
+ "
\n",
+ "
max
\n",
+ "
99926.00
\n",
+ "
120.00
\n",
+ "
46.00
\n",
+ "
409.00
\n",
+ "
478.00
\n",
+ "
1.00
\n",
+ "
\n",
+ " \n",
+ "
\n",
+ ""
+ ],
+ "text/plain": [
+ " daily_cases_clean workplaces residential retail_recreation \\\n",
+ "count 935442.00 924893.00 573705.00 591076.00 \n",
+ "mean 35.55 -19.45 4.87 0.74 \n",
+ "std 238.71 13.63 4.52 19.62 \n",
+ "min 0.00 -100.00 -29.00 -100.00 \n",
+ "25% 0.00 -25.00 2.00 -10.00 \n",
+ "50% 4.00 -18.00 4.00 0.00 \n",
+ "75% 20.00 -11.00 7.00 11.00 \n",
+ "max 99926.00 120.00 46.00 409.00 \n",
+ "\n",
+ " transit is_weekend \n",
+ "count 347338.00 935442.00 \n",
+ "mean -1.77 0.24 \n",
+ "std 33.76 0.43 \n",
+ "min -100.00 0.00 \n",
+ "25% -25.00 0.00 \n",
+ "50% -3.00 0.00 \n",
+ "75% 17.00 0.00 \n",
+ "max 478.00 1.00 "
+ ]
+ },
+ "execution_count": 61,
+ "metadata": {},
+ "output_type": "execute_result"
}
],
"source": [
- "# Setup code -- Run only once after cloning!!! \n",
- "#\n",
- "# this code downloads the data from its source to the `data/00-raw/` directory\n",
- "# if the data hasn't updated you don't need to do this again!\n",
- "\n",
- "# if you don't already have these packages (you should!) uncomment this line\n",
- "# %pip install requests tqdm\n",
- "\n",
- "import sys\n",
- "sys.path.append('./modules') # this tells python where to look for modules to import\n",
- "\n",
- "import get_data # this is where we get the function we need to download data\n",
- "\n",
- "# replace the urls and filenames in this list with your actual datafiles\n",
- "# yes you can use Google drive share links or whatever\n",
- "# format is a list of dictionaries; \n",
- "# each dict has keys of \n",
- "# 'url' where the resource is located\n",
- "# 'filename' for the local filename where it will be stored \n",
- "datafiles = [\n",
- " { 'url': 'https://raw.githubusercontent.com/fivethirtyeight/data/refs/heads/master/airline-safety/airline-safety.csv', 'filename':'airline-safety.csv'},\n",
- " { 'url': 'https://raw.githubusercontent.com/fivethirtyeight/data/refs/heads/master/bad-drivers/bad-drivers.csv', 'filename':'bad-drivers.csv'}\n",
- "]\n",
- "\n",
- "get_data.get_raw(datafiles,destination_directory='data/00-raw/')"
+ "# Summary statistics for key variables\n",
+ "key_cols = ['daily_cases_clean', 'workplaces', 'residential', 'retail_recreation', 'transit', 'is_weekend']\n",
+ "print('Summary statistics for key variables:')\n",
+ "df_clean[key_cols].describe().round(2)\n"
]
},
{
- "cell_type": "markdown",
+ "cell_type": "code",
+ "execution_count": 62,
"metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Saved to data/02-processed/covid_mobility_clean.csv\n",
+ "Final shape: (935442, 19)\n"
+ ]
+ }
+ ],
"source": [
- "### Dataset #1 \n",
- "\n",
- "Instructions: REPLACE the contents of this cell and the one below with your work, including any updates to recover points lost in your data checkpoint feedback"
+ "# Save cleaned dataset\n",
+ "os.makedirs('data/02-processed', exist_ok=True)\n",
+ "df_clean.to_csv('data/02-processed/covid_mobility_clean.csv', index=False)\n",
+ "print('Saved to data/02-processed/covid_mobility_clean.csv')\n",
+ "print('Final shape:', df_clean.shape)\n"
]
},
{
"cell_type": "code",
- "execution_count": 34,
+ "execution_count": 63,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "## YOUR CODE TO LOAD/CLEAN/TIDY/WRANGLE THE DATA GOES HERE\n",
- "## FEEL FREE TO ADD MULTIPLE CELLS PER SECTION "
+ "# EDA Plot 1: Distribution of daily_cases_clean\n",
+ "plt.figure(figsize=(8, 4))\n",
+ "df_clean[df_clean['daily_cases_clean'] < 500]['daily_cases_clean'].hist(bins=50, color='steelblue', edgecolor='black')\n",
+ "plt.xlabel('Daily Cases (filtered below 500 for readability)')\n",
+ "plt.ylabel('Frequency')\n",
+ "plt.title('Distribution of Daily COVID-19 Cases (County-Day Level)')\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n",
+ "\n",
+ "# EDA Plot 2: Average daily cases by day of week\n",
+ "dow_avg = df_clean.groupby('day_of_week')['daily_cases_clean'].mean()\n",
+ "days = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']\n",
+ "plt.figure(figsize=(7, 4))\n",
+ "plt.bar(days, dow_avg.values, color='coral', edgecolor='black')\n",
+ "plt.xlabel('Day of Week')\n",
+ "plt.ylabel('Average Daily Cases')\n",
+ "plt.title('Average Daily COVID-19 Cases by Day of Week')\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "### Dataset #2\n",
- " as above, add any more copies of this that you need to given how many datasets you have"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 35,
- "metadata": {},
- "outputs": [],
- "source": [
- "## YOUR CODE TO LOAD/CLEAN/TIDY/WRANGLE THE DATA GOES HERE\n",
- "## FEEL FREE TO ADD MULTIPLE CELLS PER SECTION "
+ "**Cleaning summary:** The raw dataset contained one row per US county per day. The `transit` and `parks` columns had the highest missingness, confirmed to be systematic - counties with missing mobility data tend to be smaller where Google had insufficient location data. We retained NaN values rather than imputing since filling them would misrepresent a true absence of data. Negative `daily_cases` values from retroactive corrections were flagged and clipped to zero in `daily_cases_clean`. No duplicate county-date rows were found. The cleaned file was saved to `data/02-processed/covid_mobility_clean.csv`.\n"
]
},
{
@@ -694,43 +1559,162 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Ethics"
+ "## Ethics \n",
+ "\n",
+ "Instructions: Keep the contents of this cell. For each item on the checklist\n",
+ "- put an X there if you've considered the item\n",
+ "- IF THE ITEM IS RELEVANT place a short paragraph after the checklist item discussing the issue.\n",
+ " \n",
+ "Items on this checklist are meant to provoke discussion among good-faith actors who take their ethical responsibilities seriously. Your teams will document these discussions and decisions for posterity using this section. You don't have to solve these problems, you just have to acknowledge any potential harm no matter how unlikely.\n",
+ "\n",
+ "Here is a [list of real world examples](https://deon.drivendata.org/examples/) for each item in the checklist that can refer to.\n",
+ "\n",
+ "[](http://deon.drivendata.org/)\n",
+ "\n",
+ "### A. Data Collection\n",
+ " - [X] **A.1 Informed consent**: If there are human subjects, have they given informed consent, where subjects affirmatively opt-in and have a clear understanding of the data uses to which they consent?\n",
+ "\n",
+ "Our analysis uses publicly available, aggregated county-level COVID-19 case counts and mobility metrics. We do not recruit participants, interact with individuals, or collect new individual-level data, so informed consent is not required for our work. However, we acknowledge that mobility metrics originate from smartphone location signals and that many users may not fully understand downstream secondary uses, even when data is anonymized and aggregated.\n",
+ "\n",
+ " - [X] **A.2 Collection bias**: Have we considered sources of bias that could be introduced during data collection and survey design and taken steps to mitigate those?\n",
+ "\n",
+ "Mobility data can underrepresent people without smartphones or stable location services (for example older adults, low income communities, unhoused populations, and some rural areas). COVID-19 case counts also reflect testing access and reporting practices that vary by county and over time. We will interpret findings as population-level associations and avoid claims that assume equal measurement quality across all counties or groups.\n",
+ "\n",
+ " - [X] **A.3 Limit PII exposure**: Have we considered ways to minimize exposure of personally identifiable information (PII) for example through anonymization or not collecting information that isn't relevant for analysis?\n",
+ "\n",
+ "Our dataset is already aggregated to county-day level and does not include direct identifiers (names, addresses, device IDs). We will not attempt any re-identification or linkage to external individual-level datasets. We will also avoid presenting results as county rankings intended to label places as good or bad.\n",
+ "\n",
+ " - [X] **A.4 Downstream bias mitigation**: Have we considered ways to enable testing downstream results for biased outcomes (e.g., collecting data on protected group status like race or gender)?\n",
+ "\n",
+ "Our results could be misused to blame specific regions or communities for COVID-19 spread or to justify punitive policy based on mobility patterns. We will present conclusions carefully, emphasizing that mobility is a proxy for many structural factors and that associations do not imply individual fault or causation. Where feasible, we will add contextual covariates (for example population, time controls, fixed effects) and include limitations that discourage stigmatizing interpretations.\n",
+ "\n",
+ "### B. Data Storage\n",
+ " - [X] **B.1 Data security**: Do we have a plan to protect and secure data (e.g., encryption at rest and in transit, access controls on internal users and third parties, access logs, and up-to-date software)?\n",
+ "\n",
+ "The data are public and aggregated, so the risk level is low, but we will still follow basic security practices: keep data in the course GitHub repo, avoid sharing raw data outside the team, and ensure devices/accounts used to access the repo are protected. We will not store any private keys or credentials in the repository.\n",
+ "\n",
+ " - [X] **B.2 Right to be forgotten**: Do we have a mechanism through which an individual can request their personal information be removed?\n",
+ "\n",
+ "Because the dataset does not contain individual-level identifiers, individuals cannot be singled out or removed from our data. We rely on the data providers' governance and opt-out mechanisms (if any). Our project does not maintain user-level records.\n",
+ "\n",
+ " - [X] **B.3 Data retention plan**: Is there a schedule or plan to delete the data after it is no longer needed?\n",
+ "\n",
+ "We will keep the dataset only as long as needed for the course and reproducibility of the final report. After the course, we can remove large raw data files from the repo while retaining derived results and code that do not contain sensitive information.\n",
+ "\n",
+ "### C. Analysis\n",
+ " - [X] **C.1 Missing perspectives**: Have we sought to address blindspots in the analysis through engagement with relevant stakeholders (e.g., checking assumptions and discussing implications with affected communities and subject matter experts)?\n",
+ "\n",
+ "Public health outcomes depend on factors not captured by mobility (healthcare access, workplace protections, housing density, local policy, and reporting practices). We will check our assumptions using public health context from reputable sources and discuss limitations so our results are not interpreted as purely behavioral explanations.\n",
+ "\n",
+ " - [X] **C.2 Dataset bias**: Have we examined the data for possible sources of bias and taken steps to mitigate or address these biases (e.g., stereotype perpetuation, confirmation bias, imbalanced classes, or omitted confounding variables)?\n",
+ "\n",
+ "We expect bias from uneven testing/reporting, time-varying policy changes, and demographic differences across counties. We will use appropriate controls (time effects, county fixed effects, lagged outcomes where relevant) and perform basic robustness checks (for example examining patterns across time periods) to reduce the chance of misleading conclusions.\n",
+ "\n",
+ " - [X] **C.3 Honest representation**: Are our visualizations, summary statistics, and reports designed to honestly represent the underlying data?\n",
+ "\n",
+ "We will avoid causal wording and report results as associations. Visualizations will clearly label units and transformations, and we will include uncertainty or variability where possible. We will not cherry-pick dates or counties to support a preferred conclusion.\n",
+ "\n",
+ " - [X] **C.4 Privacy in analysis**: Have we ensured that data with PII are not used or displayed unless necessary for the analysis?\n",
+ "\n",
+ "No PII is present in the dataset and none will be introduced. We will only analyze and display aggregated county-level metrics.\n",
+ "\n",
+ " - [X] **C.5 Auditability**: Is the process of generating the analysis well documented and reproducible if we discover issues in the future?\n",
+ "\n",
+ "All preprocessing, analysis steps, and figure generation will be documented in notebooks/scripts in the repository. This supports reproducibility and makes it easier to identify issues if errors or data problems are later discovered.\n",
+ "\n",
+ "### D. Modeling\n",
+ " - [X] **D.1 Proxy discrimination**: Have we ensured that the model does not rely on variables or proxies for variables that are unfairly discriminatory?\n",
+ "\n",
+ "Mobility measures can correlate with socioeconomic status and occupation types. We will not use models for individual-level prediction, and we will interpret results as structural and measurement patterns rather than traits of any group.\n",
+ "\n",
+ " - [X] **D.2 Fairness across groups**: Have we tested model results for fairness with respect to different affected groups (e.g., tested for disparate error rates)?\n",
+ "\n",
+ "We are not building a high-stakes decision system, but we can still check whether model residuals or fit vary across counties (for example by population size or region). If large differences appear, we will report them as limitations and avoid overgeneralizing.\n",
+ "\n",
+ " - [X] **D.3 Metric selection**: Have we considered the effects of optimizing for our defined metrics and considered additional metrics?\n",
+ "\n",
+ "We will use standard regression evaluation (for example goodness of fit and error measures) but will prioritize interpretability and stability over optimizing a single metric. We will compare multiple reasonable specifications to ensure conclusions are not driven by one modeling choice.\n",
+ "\n",
+ " - [X] **D.4 Explainability**: Can we explain in understandable terms a decision the model made in cases where a justification is needed?\n",
+ "\n",
+ "We will use interpretable models (for example linear regression with coefficients) and explain findings in plain language. We will tie interpretations directly to observed variables and note when the data do not justify a stronger claim.\n",
+ "\n",
+ " - [X] **D.5 Communicate limitations**: Have we communicated the shortcomings, limitations, and biases of the model to relevant stakeholders in ways that can be generally understood?\n",
+ "\n",
+ "Our report will include a limitations section emphasizing that associations are not causation, case reporting varies by testing/reporting practices, and mobility metrics may not represent all populations equally. We will explicitly discourage stigmatizing or punitive interpretations.\n",
+ "\n",
+ "### E. Deployment\n",
+ " - [X] **E.1 Monitoring and evaluation**: Do we have a clear plan to monitor the model and its impacts after it is deployed (e.g., performance monitoring, regular audit of sample predictions, human review of high-stakes decisions, reviewing downstream impacts of errors or low-confidence decisions, testing for concept drift)?\n",
+ "\n",
+ "We are not deploying a model in production. Still, we will sanity-check outputs (for example extreme spikes, missing dates, or inconsistent values) and document cleaning decisions to reduce errors in the final report.\n",
+ "\n",
+ " - [X] **E.2 Redress**: Have we discussed with our organization a plan for response if users are harmed by the results (e.g., how does the data science team evaluate these cases and update analysis and models to prevent future harm)?\n",
+ "\n",
+ "Because this is a course research analysis and not a deployed system, formal redress is not applicable. The main risk is misinterpretation, so we mitigate harm through careful framing, avoiding causal claims, and clearly stating limitations and appropriate uses.\n",
+ "\n",
+ " - [X] **E.3 Roll back**: Is there a way to turn off or roll back the model in production if necessary?\n",
+ "\n",
+ "Not applicable because we are not deploying a production model. If issues are found, we can update or remove analyses in the repository and clearly note changes.\n",
+ "\n",
+ " - [X] **E.4 Unintended use**: Have we taken steps to identify and prevent unintended uses and abuse of the model and do we have a plan to monitor these once the model is deployed?\n",
+ "\n",
+ "Unintended use is possible if results are used to justify surveillance or punitive restrictions targeted at particular counties. We will avoid producing county rankings meant for enforcement decisions, use neutral language, and emphasize that mobility is an imperfect proxy that should not be used for punitive action.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
- "source": [
- "Instructions: REPLACE the contents of this cell with your work, including any updates to recover points lost in your proposal feedback"
- ]
+ "source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Team Expectations "
+ "## Team Expectations\n",
+ "\n",
+ "Our team members are Nicholas Campos, Lui Gazem, Cadence Eastep, Ahgean Davis, and Chuck Davies. We have read and agree to follow the COGS108 Team Policies.\n",
+ "\n",
+ "- **Clear Communication:** We will communicate primarily through group chat and scheduled meetings. All members are expected to respond within 24 hours unless previously communicated otherwise. Important decisions will be made during meetings to ensure transparency.\n",
+ "\n",
+ "- **Accountability & Deadlines:** Each member is responsible for completing their assigned tasks on time. If someone anticipates difficulty meeting a deadline, they will notify the group as early as possible so adjustments can be made.\n",
+ "\n",
+ "- **Equal Contribution:** We expect all members to contribute meaningfully to research, coding, writing, and revisions. Work will be divided fairly, but we will support one another if someone needs assistance.\n",
+ "\n",
+ "- **Respectful Conflict Resolution:** If disagreements arise, we will address them respectfully and directly. We will focus on the work rather than personal criticism and aim to resolve issues collaboratively.\n",
+ "\n",
+ "- **Quality Control:** Before submission, all members will review the project to ensure clarity, correctness, and completeness. No section will be submitted without at least one additional team member reviewing it.\n",
+ "\n",
+ "We are committed to maintaining professionalism, consistency, and mutual respect throughout the project.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
- "source": [
- "Instructions: REPLACE the contents of this cell with your work, including any updates to recover points lost in your proposal feedback"
- ]
+ "source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Project Timeline Proposal"
+ "## Project Timeline Proposal\n",
+ "\n",
+ "Our team is currently in Week 9 of the quarter (beginning March 2nd). The final written report and presentation video are due in Week 10. We meet every Sunday at 2 PM to review progress and assign next steps. Our remaining timeline is as follows:\n",
+ "\n",
+ "| Meeting Date | Meeting Time | Completed Before Meeting | Discuss at Meeting |\n",
+ "|--------------|--------------|--------------------------|--------------------|\n",
+ "| March 2 | 2 PM | Identify key findings; Begin drafting Results section | Draft Introduction, Methods, and Results sections; Peer review draft; Revise figures; Outline presentation video structure |\n",
+ "| March 9 | 2 PM | Draft Discussion and Conclusion| Create presentation slides; Final content edits; Record presentation video |\n",
+ "| Week 10 Deadline | Before submission time | Final formatting and proofreading | Submit final written report and presentation video |\n",
+ "\n",
+ "Responsibilities for analysis, writing, visualization, and video production will be divided evenly among team members. Each major section will be reviewed by at least one additional member prior to submission to ensure clarity and accuracy.\n",
+ "\n",
+ "We do not anticipate requiring tools beyond those covered in COGS 108 (Python, pandas, visualization libraries, and standard statistical methods). If additional techniques become necessary, we will consult course materials and TA guidance before implementation.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
- "source": [
- "Instructions: Replace this with your timeline. **PLEASE UPDATE your Timeline!** No battle plan survives contact with the enemy, so make sure we understand how your plans have changed. Also if you have lost points on the previous checkpoint fix them"
- ]
+ "source": []
},
{
"cell_type": "code",
diff --git a/03-FinalProject.ipynb b/03-FinalProject.ipynb
index 262e1aa..9378af0 100644
--- a/03-FinalProject.ipynb
+++ b/03-FinalProject.ipynb
@@ -144,7 +144,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
@@ -161,7 +161,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
@@ -173,78 +173,466 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Results\n",
- "\n",
- "### Exploratory Data Analysis\n",
+ "To see the full exploratory workflow, including earlier visualizations and code development, see `02-EDACheckpoint.ipynb`. In this final report, we focus on the EDA results that most directly inform our final analyses.\n",
"\n",
- "Instructions: REPLACE the contents of this cell with your work, including any updates to recover points lost in your EDA feedback\n",
+ "The cleaned dataset contains 935,442 county-day observations from 2,814 US counties across 2021. The outcome variable, `daily_cases_clean`, is strongly right-skewed: many county-days have very few reported cases, while a smaller number of county-days in large counties during major waves have extremely high values. This skewness is visible in the distribution plot and motivates the use of `log1p(daily_cases_clean)` in later modeling so that extreme outliers do not dominate the results.\n",
"\n",
- "**NOTE** The final report should not be overly long. If this EDA step is huge and complicated and you went down a lot of dead ends, then you should do something about that. Some options... \n",
+ "The EDA also shows a clear day-of-week pattern in reported COVID-19 cases. Average reported cases are lower on Saturdays and Sundays than on weekdays, which is consistent with our hypothesis. However, this pattern likely reflects both true behavioral differences and the well-documented weekend reporting effect, in which fewer tests are processed and reported on weekends. This makes it important to formally test the weekday-weekend difference rather than relying on visual inspection alone.\n",
"\n",
- "1) remove any EDA you later decided was unecessary side quest.\n",
- "2) if the code to EDA is very long (in either lines or in time to compute) make sure you write out images of the EDA into `results/` and reload them as images here instead of recalculating it. And make a statement here like \"To see the code responsible for EDA analyses see `02-EDACheckpoint.ipynb`.\""
+ "Finally, the mobility plots suggest that higher workplace mobility tends to be associated with higher case counts, while higher residential mobility tends to be associated with lower case counts. These relationships are noisy, which is expected given substantial county-level heterogeneity in population size, policy, and pandemic timing. The temporal trend plots also show major COVID-19 waves during 2021, reinforcing the need to include a time control in our regression models."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "### First Analysis You Did - Give it a better title\n",
+ "### Analysis 1: Weekday vs. Weekend Differences in Reported COVID-19 Cases\n",
+ "\n",
+ "To formally test whether reported COVID-19 case counts differ between weekdays and weekends, we use a Mann-Whitney U test. We choose this non-parametric test instead of a t-test because `daily_cases_clean` is highly skewed and contains many low-value observations, so a normality-based test would be less appropriate. The null hypothesis is that weekday and weekend case counts come from the same distribution.\n",
+ "\n",
+ "Because statistical significance is easy to obtain with such a large dataset, we also report an effect size using Cohen’s d on log-transformed case counts. This helps distinguish between a difference that is statistically detectable and one that is substantively large.\n",
+ "Given the very large sample size (over 935,000 observations), even small differences are likely to be statistically significant."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Loaded: 935,442 rows, 19 columns\n",
+ "Date range: 2021-01-01 to 2021-12-31\n",
+ "=== Weekday vs. Weekend: Mann-Whitney U Test ===\n",
+ "Weekday N: 713,609\n",
+ "Weekend N: 221,833\n",
+ "Weekday mean cases: 39.53\n",
+ "Weekend mean cases: 22.76\n",
+ "Weekday median: 6.0\n",
+ "Weekend median: 0.0\n",
+ "U statistic: 104,369,844,388\n",
+ "p-value: 0.00e+00\n",
+ "Cohen's d (log): 0.5109\n"
+ ]
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import numpy as np\n",
+ "import matplotlib.pyplot as plt\n",
+ "import seaborn as sns\n",
+ "from scipy import stats\n",
+ "import warnings\n",
+ "warnings.filterwarnings('ignore')\n",
+ "\n",
+ "# Load directly from Google Drive (bypasses missing local file)\n",
+ "url = \"https://drive.google.com/uc?export=download&id=1hQwLJdeMyXGf2Fp0xEBXQrcj8sZsRYca\"\n",
+ "df = pd.read_csv(url)\n",
+ "df['date'] = pd.to_datetime(df['date'])\n",
"\n",
- "Some more words and stuff. Remember notebooks work best if you interleave the code that generates a result with properly annotate figures and text that puts these results into context. Every project will be different here... but some examples of analyses are: inferential statistics where you calculate p-values (like a t-test or a Monte Carlo simulation), regression models, or a machine learning model\n",
+ "# Clean: clip negatives, create log outcome\n",
+ "df['daily_cases_clean'] = df['daily_cases'].clip(lower=0)\n",
+ "df['log_daily_cases'] = np.log1p(df['daily_cases_clean'])\n",
"\n",
- "**NOTE** The final report should not be overly long. If this analysis step is huge and complicated and you went down a lot of dead ends, then you should do something about that. Some options... \n",
+ "print(f'Loaded: {df.shape[0]:,} rows, {df.shape[1]} columns')\n",
+ "print(f'Date range: {df[\"date\"].min().date()} to {df[\"date\"].max().date()}')\n",
"\n",
- "1) remove any analysis you later decided was unecessary side quest before checking in for the final hand in\n",
- "2) if the code to do an analysis is very long (in either lines or in time to compute) put it in a different file. In that file make write out results of analysis into `results/` and reload the results here instead of recalculating it. And make a statement here like \"To see the code responsible for analyses see `path/to/some-file-name.ipynb_or.py`.\""
+ "# Separate weekday and weekend samples\n",
+ "weekday_cases = df.loc[df['is_weekend'] == 0, 'daily_cases_clean'].dropna()\n",
+ "weekend_cases = df.loc[df['is_weekend'] == 1, 'daily_cases_clean'].dropna()\n",
+ "\n",
+ "# Mann-Whitney U test (one-sided: weekday > weekend)\n",
+ "u_stat, p_value = stats.mannwhitneyu(weekday_cases, weekend_cases, alternative='greater')\n",
+ "\n",
+ "# Cohen's d on log-transformed values\n",
+ "log_weekday = np.log1p(weekday_cases)\n",
+ "log_weekend = np.log1p(weekend_cases)\n",
+ "pooled_std = np.sqrt(\n",
+ " (log_weekday.std()**2 * (len(log_weekday)-1) + log_weekend.std()**2 * (len(log_weekend)-1))\n",
+ " / (len(log_weekday) + len(log_weekend) - 2)\n",
+ ")\n",
+ "cohens_d = (log_weekday.mean() - log_weekend.mean()) / pooled_std\n",
+ "\n",
+ "print('=== Weekday vs. Weekend: Mann-Whitney U Test ===')\n",
+ "print(f'Weekday N: {len(weekday_cases):>10,}')\n",
+ "print(f'Weekend N: {len(weekend_cases):>10,}')\n",
+ "print(f'Weekday mean cases: {weekday_cases.mean():>10.2f}')\n",
+ "print(f'Weekend mean cases: {weekend_cases.mean():>10.2f}')\n",
+ "print(f'Weekday median: {weekday_cases.median():>10.1f}')\n",
+ "print(f'Weekend median: {weekend_cases.median():>10.1f}')\n",
+ "print(f'U statistic: {u_stat:>10,.0f}')\n",
+ "print(f'p-value: {p_value:.2e}')\n",
+ "print(f\"Cohen's d (log): {cohens_d:>10.4f}\")"
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "# Figure: Weekday vs Weekend distributions\n",
+ "fig, axes = plt.subplots(1, 2, figsize=(11, 5))\n",
+ "\n",
+ "plot_data = df[['is_weekend', 'log_daily_cases']].dropna().copy()\n",
+ "plot_data['Day Type'] = plot_data['is_weekend'].map({0.0: 'Weekday', 1.0: 'Weekend'})\n",
+ "\n",
+ "# Violin plot\n",
+ "sns.violinplot(\n",
+ " data=plot_data, x='Day Type', y='log_daily_cases',\n",
+ " palette={'Weekday': '#2980b9', 'Weekend': '#c0392b'},\n",
+ " inner='quartile', ax=axes[0], cut=0\n",
+ ")\n",
+ "axes[0].set_xlabel('Day Type', fontsize=11)\n",
+ "axes[0].set_ylabel('log(1 + Daily Cases)', fontsize=11)\n",
+ "axes[0].set_title('A. Log Daily Cases by Day Type', fontsize=11)\n",
+ "\n",
+ "# Boxplot by day of week\n",
+ "dow_plot = df[['day_of_week', 'log_daily_cases']].dropna().copy()\n",
+ "day_labels_map = {0:'Mon', 1:'Tue', 2:'Wed', 3:'Thu', 4:'Fri', 5:'Sat', 6:'Sun'}\n",
+ "dow_plot['Day'] = dow_plot['day_of_week'].map(day_labels_map)\n",
+ "day_order = ['Mon','Tue','Wed','Thu','Fri','Sat','Sun']\n",
+ "box_colors = ['#2980b9']*5 + ['#c0392b']*2\n",
+ "\n",
+ "bp = axes[1].boxplot(\n",
+ " [dow_plot.loc[dow_plot['Day']==d, 'log_daily_cases'].values for d in day_order],\n",
+ " labels=day_order, patch_artist=True, showfliers=False,\n",
+ " medianprops=dict(color='black', linewidth=1.5)\n",
+ ")\n",
+ "for patch, color in zip(bp['boxes'], box_colors):\n",
+ " patch.set_facecolor(color)\n",
+ " patch.set_alpha(0.7)\n",
+ "axes[1].set_xlabel('Day of Week', fontsize=11)\n",
+ "axes[1].set_ylabel('log(1 + Daily Cases)', fontsize=11)\n",
+ "axes[1].set_title('B. Log Daily Cases by Day of Week', fontsize=11)\n",
+ "\n",
+ "plt.suptitle(\n",
+ " 'Figure 1. Weekday vs. Weekend Distribution of Log-Transformed Daily COVID-19 Cases\\n'\n",
+ " 'Blue = weekdays, red = weekends. Weekday medians are consistently higher across all five weekdays.\\n'\n",
+ " 'Mann-Whitney U test: p < 0.001. Outliers suppressed for readability.',\n",
+ " fontsize=9, y=-0.05, ha='center'\n",
+ ")\n",
+ "plt.tight_layout()\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
"metadata": {},
- "outputs": [],
"source": [
- "## YOUR CODE HERE\n",
- "## FEEL FREE TO ADD MULTIPLE CELLS PER SECTION"
+ "Figure 1 caption. Boxplots of log-transformed daily COVID-19 case counts by day of week. Weekdays are shown in blue and weekends in red. Outliers are suppressed for readability.\n",
+ "\n",
+ "The Mann-Whitney U test indicates that weekday case counts are significantly higher than weekend case counts (`p < 0.001`). This supports our hypothesis that reported daily cases vary systematically across the week. The effect size (Cohen’s d ≈ 0.51) indicates a moderate difference between weekday and weekend case distributions. However, day-of-week alone still explains only part of the overall variation in county-level case counts. This is consistent with the broader context of the pandemic, where population size, regional waves, testing and reporting practices, and local conditions are also major drivers of case counts.\n",
+ "\n",
+ "Importantly, this result should not be interpreted as purely behavioral. The weekend reporting effect likely contributes to the lower weekend values we observe, meaning that the weekday-weekend difference reflects both real social patterns and administrative reporting patterns."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "### Second Analysis You Did - Give it a better title\n",
+ "### Analysis 2: Regression Models of Mobility and Daily COVID-19 Cases\n",
"\n",
- "Some more words and stuff. Remember notebooks work best if you interleave the code that generates a result with properly annotate figures and text that puts these results into context. Every project will be different here... but some examples of analyses are: inferential statistics where you calculate p-values (like a t-test or a Monte Carlo simulation), regression models, or a machine learning model"
+ "To quantify the relationship between mobility behavior and COVID-19 case counts while controlling for day-of-week and time, we fit a series of Ordinary Least Squares (OLS) regression models with `log_daily_cases` as the outcome variable. We use the log-transformed outcome because the raw case counts are highly skewed.\n",
+ "\n",
+ "We compare three models of increasing complexity. Model 1 includes only `is_weekend` as a baseline test of weekday-weekend reporting effects. Model 2 adds the two mobility variables most central to our hypothesis: `workplaces` and `residential`. Model 3 adds `retail_recreation`, `transit`, `is_holiday`, and a week-index time trend. We evaluate model performance using `R²`, Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). These metrics help show both how much variance the model explains and how far predictions tend to be from observed values."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 13,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " Model N R² MAE (log) RMSE (log) MAE (cases)\n",
+ " Model 1: Weekend only 935,442 0.0451 1.3904 1.6655 34.5\n",
+ "Model 2: Weekend + key mobility 573,705 0.0964 1.4209 1.7128 51.8\n",
+ " Model 3: Full specification 289,751 0.1837 1.4751 1.7888 82.4\n",
+ "\n",
+ "Model 3 Coefficients:\n",
+ " is_weekend : -1.0677\n",
+ " workplaces : +0.0151\n",
+ " residential : +0.0953\n",
+ " retail_recreation : -0.0122\n",
+ " transit : -0.0076\n",
+ " is_holiday : -1.8252\n",
+ " week_index : +0.0062\n"
+ ]
+ }
+ ],
+ "source": [
+ "from sklearn.linear_model import LinearRegression\n",
+ "from sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error\n",
+ "\n",
+ "# Create week index time trend\n",
+ "df['week_index'] = ((df['date'] - df['date'].min()).dt.days // 7).astype(int)\n",
+ "\n",
+ "def fit_ols(feature_cols, label):\n",
+ " \"\"\"Fit OLS regression on complete cases for given features.\n",
+ " \n",
+ " Parameters\n",
+ " ----------\n",
+ " feature_cols : list of str\n",
+ " Predictor column names.\n",
+ " label : str\n",
+ " Model label for display.\n",
+ " \n",
+ " Returns\n",
+ " -------\n",
+ " dict with model, metrics, and coefficients.\n",
+ " \"\"\"\n",
+ " subset = df[feature_cols + ['log_daily_cases']].dropna()\n",
+ " X = subset[feature_cols].values\n",
+ " y = subset['log_daily_cases'].values\n",
+ " model = LinearRegression()\n",
+ " model.fit(X, y)\n",
+ " y_pred = model.predict(X)\n",
+ " mae_cases = np.mean(np.abs(np.expm1(y) - np.expm1(y_pred)))\n",
+ " return {\n",
+ " 'label': label,\n",
+ " 'n': len(y),\n",
+ " 'r2': r2_score(y, y_pred),\n",
+ " 'mae_log': mean_absolute_error(y, y_pred),\n",
+ " 'rmse_log': np.sqrt(mean_squared_error(y, y_pred)),\n",
+ " 'mae_cases': mae_cases,\n",
+ " 'coefs': dict(zip(feature_cols, model.coef_)),\n",
+ " 'intercept': model.intercept_,\n",
+ " 'features': feature_cols\n",
+ " }\n",
+ "\n",
+ "m1 = fit_ols(['is_weekend'], 'Model 1: Weekend only')\n",
+ "m2 = fit_ols(['is_weekend', 'workplaces', 'residential'], 'Model 2: Weekend + key mobility')\n",
+ "m3 = fit_ols(\n",
+ " ['is_weekend', 'workplaces', 'residential', 'retail_recreation', 'transit', 'is_holiday', 'week_index'],\n",
+ " 'Model 3: Full specification'\n",
+ ")\n",
+ "\n",
+ "# Print results table\n",
+ "results_table = pd.DataFrame([\n",
+ " {\n",
+ " 'Model': m['label'],\n",
+ " 'N': f\"{m['n']:,}\",\n",
+ " 'R²': f\"{m['r2']:.4f}\",\n",
+ " 'MAE (log)': f\"{m['mae_log']:.4f}\",\n",
+ " 'RMSE (log)': f\"{m['rmse_log']:.4f}\",\n",
+ " 'MAE (cases)': f\"{m['mae_cases']:.1f}\"\n",
+ " }\n",
+ " for m in [m1, m2, m3]\n",
+ "])\n",
+ "print(results_table.to_string(index=False))\n",
+ "print()\n",
+ "print('Model 3 Coefficients:')\n",
+ "for feat, coef in m3['coefs'].items():\n",
+ " print(f' {feat:25s}: {coef:+.4f}')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "## YOUR CODE HERE\n",
- "## FEEL FREE TO ADD MULTIPLE CELLS PER SECTION"
+ "# Figure: Model comparison and coefficient plot\n",
+ "fig, axes = plt.subplots(1, 2, figsize=(13, 5))\n",
+ "\n",
+ "# Panel A: R² bar chart\n",
+ "model_labels = ['Model 1\\n(weekend only)', 'Model 2\\n(+ key mobility)', 'Model 3\\n(full)']\n",
+ "r2_vals = [m1['r2'], m2['r2'], m3['r2']]\n",
+ "axes[0].bar(model_labels, r2_vals, color=['#bdc3c7','#85c1e9','#2980b9'], edgecolor='black', linewidth=0.8)\n",
+ "for i, v in enumerate(r2_vals):\n",
+ " axes[0].text(i, v + 0.0005, f'{v:.4f}', ha='center', va='bottom', fontsize=10, fontweight='bold')\n",
+ "axes[0].set_ylabel('R² (proportion of variance explained)', fontsize=11)\n",
+ "axes[0].set_title('A. Model R² Comparison', fontsize=11)\n",
+ "axes[0].set_ylim(0, max(r2_vals) * 1.35)\n",
+ "\n",
+ "# Panel B: Coefficient plot for Model 3\n",
+ "feat_labels = {\n",
+ " 'is_weekend': 'Weekend (vs. Weekday)',\n",
+ " 'workplaces': 'Workplace Mobility',\n",
+ " 'residential': 'Residential Mobility',\n",
+ " 'retail_recreation': 'Retail & Recreation',\n",
+ " 'transit': 'Transit Mobility',\n",
+ " 'is_holiday': 'Holiday',\n",
+ " 'week_index': 'Week Index (time trend)'\n",
+ "}\n",
+ "coef_vals = [m3['coefs'][f] for f in m3['features']]\n",
+ "labels = [feat_labels[f] for f in m3['features']]\n",
+ "colors_coef = ['#c0392b' if c < 0 else '#27ae60' for c in coef_vals]\n",
+ "\n",
+ "axes[1].barh(range(len(labels)), coef_vals, color=colors_coef, edgecolor='black', linewidth=0.6, alpha=0.85)\n",
+ "axes[1].axvline(0, color='black', linewidth=1.0)\n",
+ "axes[1].set_yticks(range(len(labels)))\n",
+ "axes[1].set_yticklabels(labels, fontsize=10)\n",
+ "axes[1].set_xlabel('OLS Coefficient (outcome: log daily cases)', fontsize=11)\n",
+ "axes[1].set_title('B. Model 3 Coefficient Estimates\\n(green = positive association, red = negative)', fontsize=11)\n",
+ "\n",
+ "plt.suptitle(\n",
+ " 'Figure 2. OLS Regression Results: Mobility and Day-of-Week Predictors of Log Daily COVID-19 Cases\\n'\n",
+ " 'Model 3 includes all mobility variables, holiday flag, and week-index time trend.\\n'\n",
+ " 'All directional associations are consistent with our hypothesis; overall R² remains modest.',\n",
+ " fontsize=9, y=-0.05, ha='center'\n",
+ ")\n",
+ "plt.tight_layout()\n",
+ "plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "### ETC AD NASEUM\n",
+ "Figure 2A caption. Comparison of `R²` values across the three regression models.\n",
+ "\n",
+ "Figure 2B caption. Coefficient estimates from the full regression specification. Positive coefficients indicate a positive association with log daily cases, while negative coefficients indicate a negative association.\n",
"\n",
- "Some more words and stuff. Remember notebooks work best if you interleave the code that generates a result with properly annotate figures and text that puts these results into context."
+ "The regression results are mixed. Across all model specifications, the coefficient for `is_weekend` is negative, indicating that weekends are associated with lower reported case counts even after accounting for mobility and time. The coefficient for `workplaces` is positive, which is consistent with our hypothesis that greater workplace mobility is associated with higher COVID-19 case counts. However, the coefficient for `residential` is also positive in the full model, which does not match our original hypothesis. This suggests that once additional predictors and time trends are included, residential mobility may be capturing broader pandemic conditions rather than simply reflecting protective stay-at-home behavior.\n",
+ "\n",
+ "The performance metrics also help contextualize these findings. `R²` shows the proportion of variance explained by each model, while MAE and RMSE summarize average prediction error on the log-transformed outcome scale. Although the fuller models improve on the weekend-only baseline, the `R²` values remain modest. This indicates that mobility and day-of-week patterns explain only part of the variation in county-level COVID-19 cases. That result is reasonable, since many important influences on case counts—such as county population size, policy differences, vaccination levels, and variant dynamics—are not fully captured in this dataset."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Additional Temporal Context\n",
+ "The temporal trends observed in `02-EDACheckpoint.ipynb` provide additional context for the regression results. The rolling-average case plots show major waves during January and late summer/fall of 2021, which helps explain why raw mobility-case relationships are noisy when viewed without time controls. Including a week-index trend in the regression model therefore improves the interpretation of mobility effects by accounting for broad pandemic-wide shifts over time."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 15,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "=== Lagged Correlations: Weekly Workplaces vs. Avg Cases ===\n",
+ " Lag 0 weeks: r = -0.366, p = 0.007\n",
+ " Lag 1 weeks: r = -0.386, p = 0.005\n",
+ " Lag 2 weeks: r = +0.047, p = 0.743\n",
+ " Lag 3 weeks: r = +0.062, p = 0.668\n",
+ "\n",
+ "=== Lagged Correlations: Weekly Residential vs. Avg Cases ===\n",
+ " Lag 0 weeks: r = +0.538, p = 0.000\n",
+ " Lag 1 weeks: r = +0.421, p = 0.002\n",
+ " Lag 2 weeks: r = +0.182, p = 0.202\n",
+ " Lag 3 weeks: r = +0.100, p = 0.490\n"
+ ]
+ }
+ ],
+ "source": [
+ "# National weekly averages\n",
+ "df['week'] = df['date'].dt.to_period('W')\n",
+ "weekly = df.groupby('week').agg(\n",
+ " avg_cases=('daily_cases_clean', 'mean'),\n",
+ " avg_workplaces=('workplaces', 'mean'),\n",
+ " avg_residential=('residential', 'mean'),\n",
+ ").reset_index()\n",
+ "weekly['week_start'] = weekly['week'].apply(lambda x: x.start_time)\n",
+ "weekly = weekly.sort_values('week_start').reset_index(drop=True)\n",
+ "\n",
+ "print('=== Lagged Correlations: Weekly Workplaces vs. Avg Cases ===')\n",
+ "for lag in range(0, 4):\n",
+ " x = weekly['avg_workplaces'].iloc[:len(weekly)-lag] if lag > 0 else weekly['avg_workplaces']\n",
+ " y = weekly['avg_cases'].iloc[lag:] if lag > 0 else weekly['avg_cases']\n",
+ " r, p = stats.pearsonr(x, y)\n",
+ " print(f' Lag {lag} weeks: r = {r:+.3f}, p = {p:.3f}')\n",
+ "\n",
+ "print()\n",
+ "print('=== Lagged Correlations: Weekly Residential vs. Avg Cases ===')\n",
+ "for lag in range(0, 4):\n",
+ " x = weekly['avg_residential'].iloc[:len(weekly)-lag] if lag > 0 else weekly['avg_residential']\n",
+ " y = weekly['avg_cases'].iloc[lag:] if lag > 0 else weekly['avg_cases']\n",
+ " r, p = stats.pearsonr(x, y)\n",
+ " print(f' Lag {lag} weeks: r = {r:+.3f}, p = {p:.3f}')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "## YOUR CODE HERE\n",
- "## FEEL FREE TO ADD MULTIPLE CELLS PER SECTION"
+ "# Figure: National time trends\n",
+ "daily_national = df.groupby('date')['daily_cases_clean'].mean().reset_index()\n",
+ "daily_national['rolling'] = daily_national['daily_cases_clean'].rolling(7, center=True).mean()\n",
+ "\n",
+ "fig, ax1 = plt.subplots(figsize=(13, 5))\n",
+ "\n",
+ "ax1.fill_between(daily_national['date'], daily_national['rolling'], alpha=0.2, color='steelblue')\n",
+ "ax1.plot(daily_national['date'], daily_national['rolling'], color='steelblue', linewidth=2,\n",
+ " label='Avg daily cases (7-day rolling mean)')\n",
+ "ax1.set_ylabel('Average Daily Cases per County\\n(7-day rolling mean)', fontsize=11, color='steelblue')\n",
+ "ax1.tick_params(axis='y', labelcolor='steelblue')\n",
+ "ax1.set_xlabel('Date (2021)', fontsize=11)\n",
+ "\n",
+ "ax2 = ax1.twinx()\n",
+ "ax2.plot(weekly['week_start'], weekly['avg_workplaces'], color='#e74c3c', linewidth=1.8,\n",
+ " label='Avg workplace mobility')\n",
+ "ax2.plot(weekly['week_start'], weekly['avg_residential'], color='#27ae60', linewidth=1.8,\n",
+ " linestyle='--', label='Avg residential mobility')\n",
+ "ax2.axhline(0, color='gray', linewidth=0.8, linestyle=':')\n",
+ "ax2.set_ylabel('Mobility Change from Baseline (%)', fontsize=11)\n",
+ "\n",
+ "lines1, labels1 = ax1.get_legend_handles_labels()\n",
+ "lines2, labels2 = ax2.get_legend_handles_labels()\n",
+ "ax1.legend(lines1 + lines2, labels1 + labels2, loc='upper right', fontsize=9)\n",
+ "ax1.set_title('National Weekly Trends: COVID-19 Cases and Mobility (2021)', fontsize=12)\n",
+ "\n",
+ "plt.suptitle(\n",
+ " 'Figure 3. National Average Daily COVID-19 Cases (blue, left axis) and Weekly Mobility Changes (right axis) Across 2021.\\n'\n",
+ " 'The January wave and Delta wave are visible. Workplace mobility recovery in summer 2021 partially coincides with rising Delta wave cases.',\n",
+ " fontsize=9, y=-0.05, ha='center'\n",
+ ")\n",
+ "plt.tight_layout()\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Figure 3 caption. National average daily COVID-19 cases are shown with a 7-day rolling mean, alongside weekly average workplace and residential mobility. These trends provide temporal context for the mobility-case relationship observed in the regression models.\n",
+ "\n",
+ "The lagged correlation results suggest that the relationship between mobility and case counts is more complex than our original hypothesis implied. Workplace mobility shows negative short-lag correlations with average case counts, while residential mobility shows positive short-lag correlations. This likely reflects broader pandemic dynamics: during major surges, workplace mobility decreased and residential mobility increased as people stayed home more. These results reinforce that mobility measures can reflect both behavioral responses to rising cases and potential drivers of transmission, making simple causal interpretation difficult."
]
},
{
@@ -275,6 +663,27 @@
"5. Discuss the limitations of your work. e.g, is the dataset a biased sample? would more data make you more confident? would longitudinal instead of cross-sectional data helpful? is there evidence that the phenomena you're investigating may change over time? etc\n",
"6. If you would like, you can also say what the next steps in this investigation might be if the work was to continue outside this class\n"
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
}
],
"metadata": {
@@ -293,7 +702,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.11.13"
+ "version": "3.11.9"
}
},
"nbformat": 4,
diff --git a/modules/eda_utils.py b/modules/eda_utils.py
index 9d467f3..df31d16 100644
--- a/modules/eda_utils.py
+++ b/modules/eda_utils.py
@@ -48,4 +48,4 @@ def plot_time_trend(df):
daily_avg.rolling(7).mean().plot()
plt.title('7-Day Rolling Average COVID Cases')
plt.ylabel('Average Daily Cases')
- plt.show()
\ No newline at end of file
+ plt.show()