output

pdf_document	html_document
default	default

About this Repository

This repository contains code and documentation relating to a manuscript titled, 'The (In)Stability of Work-based Risk Factors for Drinking Over Time: Insights from a Two-Wave Analysis of Emerging Adults Over the Course of Initial Career Entry'.

The code in this repository was utilized to prepare and analyze data collected from the College-to-Work Transition Study (C2W). Focused on young adults transitioning from college to the work force, C2W involved a multi-year data collection effort where seniors in their final quarter/semester before graduation from universities located in the Pacific Northwest, Midwest, Southeast, and Northeast parts of the USA were invited to participate. Analyses involved investigating hypotheses relating to the time varying change in alcohol use between the Early Onboarding Phase and the Late Onboarding Phase, moments in time corresponding to when young adults first engage in full time work and at twelve months immediately afterwards, respectively. Specifically, the following were used as outcomes (i.e., dependent variables):

Number of days within the past month with any heavy drinking (HED): HED was assessed with a one-item measure. Participants were asked, 'How often in the past month did you drink more than 4 (if you are a woman) or 5 (if you are a man) standard drinks in a single day?' Responses for each item ranged from 0 (= None in the past month) to 30 (= 30 days in the past month).
Alcohol Related Problems: Alcohol Related Problems was measured using the Rutgers Alcohol Problem Index (RAPI). RAPI was assessed with a 19-item measure drawn from a broader 23-item measure developed by White & LaBouvie (White & LaBouvie, 1989). Each item asked participants to rate on a scale of 1 (never) to 5 (More than 10 times) the frequency of a given event happening in the past month while they were drinking or because of their alcohol use. Items included, 'Got into fights, acted bad, or did mean things', 'Felt that you needed more alcohol than you used to have in order to get the same effect', and 'Felt that you had a problem with alcohol'. Responses for each item were rescaled to range from 0 (never) to 4 (More than 10 times) and then summed to obtain an overall score.

1. Record of R packages utilized

File	Brief Description
renv.lock	Records the collection of R packages and specific versions utilized. The R package `renv` was used to take a snapshot of the software version numbers.

2. Constructing datasets used in analyses

Code within the `construct-dataset' folder

File	Brief Description
data-pipeline.R	Documents sequence in which the scripts within the `construct-dataset' folder were run.
construct-variables-baseline.R	Construct variables using raw data collected at baseline (i.e., the 'baseline dataset'); raw data is in tabular format.
construct-variables-screener.R	Construct variables raw data collected at screening (i.e., the 'screener dataset'); raw data is in tabular format.
construct-variables-wnw.R	Construct variables using raw data collected after baseline (i.e., the 'wnw dataset'); raw data is in tabular format.
reshape-data.R	Identify rows within the wnw dataset to associate with the first time an individual reports to be full-time employed and twelve months after that time. In addition, apply exclusion criteria (e.g., exclude individuals who are not full time employed at both of these timepoints).

3. Exploratory data analyses

Code within the `explore-dataset' folder

File	Brief Description
display-possible-range-of-values.Rmd	Displays the range of possible range of values variables can take on. Results are displayed in display-possible-range-of-values.pdf
display-actual-range-of-values.Rmd	Displays the range of actual range of values variables can take on. Results are displayed in display-actual-range-of-values.pdf
display-correlation-matrix.Rmd	Displays the correlation among variables . Results are displayed in display-correlation-matrix.pdf; to enable results to fit in one page, variable names are not displayed. The numbers 1, 2, ..., 23 labeling the rows and columns in this file correspond to the variables listed in order from the 1st, 2nd, ..., 23rd row in the file display-actual-range-of-values.pdf
test-homogeneity-RAPI.Rmd and test-homogeneity-HED.Rmd	Conduct Chi-squared tests (for binary or count variables) or Kolmogorov-Smirnov tests (for continuous variables) to test for homogeneity. Results are displayed in test-homogeneity-RAPI.pdf and test-homogeneity-HED.pdf

3. Testing hypotheses

Code within the `analysis' folder

Subfolder	Brief Description
Hypothesis H01a and Hypothesis 01b	Code and results relating to H1a and H1b
Hypothesis H01c	Code and results relating to H1c
Hypothesis H02	Code and results relating to H2
Hypothesis H03	Code and results relating to H3
Hypothesis H04	Code and results relating to H4

Other

File	Brief Description
analysis-utils.R	A collection of functions for obtaining summary statistics, estimating moderated effects, and formatting output.
word-documents	A folder containing results displayed in camera-ready format.

jamieyap/College-to-Work-Transition-Study