CHL 640 Week 5 Data Collection and Analysis Plan Example

From Weigh-Ins to Data Parties: A Data Collection and Analysis Plan for the Walk and Eat Well Evaluation

[Student Name]

University of Phoenix

CHL/640: Evaluating Community Health Initiatives

Week 5 Assignment

[Instructor Name]

[Date]

The initiative, protocols, schedules, staffing, analyses and community sessions are composites written for a model paper modeled on Yakima, Washington; research findings come from the sources cited.

In the first month of the evaluation, the health district's evaluator reviewed weight records from three lifestyle groups and found a problem: one promotora weighed participants with shoes on, another without and a third rounded to the nearest pound. Small inconsistencies like these could hide or exaggerate the program's effects. This paper sets out the data collection and analysis plan that fixed them.

Organizing by Data Source

The plan is organized by source, matching the indicator table: program records, weigh-ins, participant surveys, step counts, store audits and sales, path observations and interviews. For each, it defines who collects data, when, how and how quality is checked.

Program Records

Promotoras record attendance on sign-in sheets and enter them into the program database each week. The coordinator reviews completeness every Friday and follows up with any group missing records. Missing entries are resolved within a week.

Weigh-Ins

The new protocol requires calibrated digital scales, checked monthly with a standard weight; participants weighed without shoes and heavy clothing, in private; and weights recorded to one decimal place. Each promotora was retrained, and the lead promotora observes one weigh-in per group per month.

Surveys

Surveys at enrollment, six and twelve months are offered in Spanish or English, on paper or read aloud by a promotora not leading that participant's group, to reduce pressure to give favorable answers. Items were pretested with residents.

Step Counts

Participants wear pedometers for two weeks at each time point and record daily steps on a picture log. Promotoras check logs weekly and replace lost devices, and participants who forget to wear pedometers for several days extend their two-week window.

Store Audits and Sales

The coordinator audits each store monthly with a checklist of the 20 agreed items, noting presence, placement and price. Owners provide monthly produce sales totals under data agreements, and two comparison stores outside the program provide the same totals for the time series.

Path Observations

Trained observers count path users on set evenings using SOPARC, a park observation tool that was tested with 16,244 individuals in 165 park areas, with correlations between independent observers of 0.99 on counts of users while classifications of users' age, background and how vigorously they moved also reached acceptable agreement (McKenzie et al., 2006). The team adapted its protocol, training observers until paired counts agreed closely, and repeats paired checks each quarter.

Interviews

Trained interviewers conduct exit interviews in the participant's preferred language, in a place the participant chooses, using a semistructured guide. Interviews are recorded with consent, transcribed and translated where needed, with a bilingual team member checking translations.

Training Data Collectors

Everyone who collects data completes training suited to the task: promotoras on weighing, surveys and step logs; observers on path counts; interviewers on the guide, probing and neutrality. Training includes practice with feedback, and collectors are certified before working alone. A short refresher follows any change in protocol, and new staff are trained before their first data collection.

Consent and Confidentiality

Participants give written or verbal consent in their language, may skip any question and may withdraw at any time. Names are separated from data using study numbers. Immigration status is never asked or recorded. Access to records with names is limited to two people, the evaluator and the coordinator, and every report presents grouped results.

Quantitative Analysis

Analyses are specified before data arrive. Reach and attendance are described with counts and percentages by language, age, sex and neighborhood. Weight change and activity are compared between participants and the matched comparison group using change scores and regression models adjusting for baseline differences. Store sales and path counts are analyzed with segmented regression, following the time series design, with adjustments for season. Results are reported with confidence intervals.

Missing Data

Participants who stop attending will be contacted for twelve-month weights where possible. Analyses will compare completers with all enrollees and use methods that account for missing data, reporting both so readers can judge the effect of dropout.

Qualitative Analysis

Interviews are analyzed using thematic analysis, as Braun and Clarke lay it out. Analysts first immerse themselves in the transcripts and tag meaningful passages, then group the tags into candidate themes, test those themes against the data, settle what each theme means and finally write it up with supporting quotes; the authors present the approach as flexible rather than tied to one theory (Braun & Clarke, 2006). Two coders, including a resident trained for the task, code independently and then compare. A resident coder hears things in an interview that an outside analyst may miss.

Analysis Software and Documentation

Quantitative analyses are run in a standard statistical package, with code saved and reviewed by the university reviewer so results can be reproduced. Interview coding uses a simple shared spreadsheet rather than specialized software, so the resident coder can work alongside the evaluator. Every analytic decision, such as how outliers are handled, is logged.

Integrating Numbers and Stories

Results are brought together in joint displays: a table placing each quantitative finding next to related interview themes. For example, if attendance drops in September, interview themes about harvest work appear beside the attendance chart.

Community Data Parties

Before conclusions are written, results go back to the community. Principles of community-based research call for involving community members in all phases of research, including interpretation, and for integrating knowledge with action (Israel et al., 1998). Twice a year, the team holds data parties at the community center where residents, promotoras and partners review charts and quotes, discuss what they mean and suggest explanations and changes.

Quality Control Summary

Quality control includes written protocols, training with practice, agreement checks for weights, observations and coding, weekly completeness checks and monthly scale calibration.

Handling Unexpected Findings

If results surprise the team, such as a store's sales falling after the program began, the plan calls for checking data quality first, then asking the people involved what happened before drawing conclusions. Surprises are reported, not hidden, because they often teach the most.

Calendar

Enrollment surveys and weights at intake; attendance weekly; store audits and sales monthly; path observations on set evenings monthly; six- and twelve-month surveys, weights and step counts; exit interviews after each cohort; data parties every six months; and a full analysis at month eighteen.

Roles

The evaluator leads analysis and quality control. The coordinator manages records and store data. Promotoras collect weights, surveys and step logs. Trained residents conduct path observations and co-code interviews. The university reviewer checks the analysis plan and final report.

Conclusion

The data plan turns the evaluation design into consistent procedures: protocols that fixed three different weighing habits, trained observers using a validated system, private and culturally appropriate surveys and interviews and analyses specified in advance. Thematic analysis gives structure to interviews, joint displays connect numbers and stories and data parties grounded in community-based research put residents at the center of interpreting what the results mean.

References

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. https://doi.org/10.1191/1478088706qp063oa

Israel, B. A., Schulz, A. J., Parker, E. A., & Becker, A. B. (1998). Review of community-based research: Assessing partnership approaches to improve public health. Annual Review of Public Health, 19, 173-202. https://doi.org/10.1146/annurev.publhealth.19.1.173

McKenzie, T. L., Cohen, D. A., Sehgal, A., Williamson, S., & Golinelli, D. (2006). System for Observing Play and Recreation in Communities (SOPARC): Reliability and feasibility measures. Journal of Physical Activity and Health, 3(Suppl. 1), S208-S222. https://doi.org/10.1123/jpah.3.s1.s208