| Course | PHLT 8044 Advanced Program Evaluation |
|---|---|
| Week | Week 6 |
| Paper type | Design memo |
| Length | About 810 words, 4 pages |
| Format | Memo, APA 7 citations |
| School | Walden University |
| Program | PhD in Public Health |
| Updated | October 2026 |
Free sample paper for PHLT 8044 Week 6
Memorandum
To: State School Water Access Program Manager and Evaluation Advisory Group
From: Lead Evaluator (DrPH student)
Date: Month Day, Year
Re: The comparison the outcome evaluation will use, chosen before any outcome data are seen, and its known weaknesses
The Decision
This memo commits the outcome evaluation to one design: a difference-in-differences comparison of the 30 middle schools receiving bottle-filling stations this fall against the 40 schools on the waiting list for the following year. The primary outcome is cafeteria sugary drink purchases per enrolled student per week, from district food service sales records, measured for the two school years before installation and the school year after. The comparison was chosen now, before any post-installation data exist, so that it cannot be adjusted to favor a result. Its weaknesses are written down here for the same reason, so that no one can later describe a weakness discovered in the results as a surprise the design could not have anticipated.
Why This Comparison
Three alternatives were considered and rejected. A simple before-and-after comparison of program schools would attribute any change to the program, including changes caused by new district beverage policies, price changes or hot weather. A comparison of students who carry bottles with those who do not, proposed in discussion, would confuse the program's effect with the characteristics of students who choose to carry bottles. A randomized assignment of schools to install now or later would be strongest, but installation order was set last spring by facilities schedules and principals' applications, and the program manager could not change it.
The waiting-list comparison makes the most of the program's phased rollout. Both groups applied for the program, so they share whatever motivation led principals to apply. Using each school's own baseline cancels out fixed traits such as enrollment, income mix and cafeteria layout. What remains is the difference in how purchases changed in program schools compared with how they changed in waiting-list schools over the same period. Wing et al. (2018) describe this design as well suited to policies introduced at different times in different places, provided its key assumption is checked.
The Key Assumption and How It Will Be Checked
The design assumes that, without the program, purchases in program and waiting-list schools would have changed by similar amounts, the parallel trends assumption. It cannot be proven, but it can be examined. Sales records exist for the two school years before installation, so the evaluation will compare the trend in each group across those years before seeing any post-installation data. If the trends diverge noticeably, the memo's design will be revised and the revision documented before outcomes are analyzed.
Threats Left Standing
Shadish et al. (2002) organize threats to validity into categories, and several remain in this design.
The most serious remaining threat is history interacting with selection. Program and waiting-list schools are not spread evenly across districts, so a district-wide change in beverage contracts could fall mostly on one group. The handling plan cannot eliminate this threat; it can only detect large shifts and test whether results hold when affected districts are removed.
Table 1
Remaining Threats to Validity and How Each Will Be Handled
| Threat | How it could bias the estimate | Handling | |
|---|---|---|
| Selection by history interaction | An event affecting one group, such as a new beverage contract in a district with many program schools | Record district policy changes; sensitivity analysis excluding affected districts | |
| Instrumentation | A change in cafeteria point-of-sale systems that alters how drinks are recorded | Confirm item codes with food service before and after | |
| Diffusion | Waiting-list schools adding fountains or bottles on their own after seeing the program | Ask waiting-list principals about water changes each term | |
| Regression to the mean | Program schools chosen partly because purchases were unusually high | Use two baseline years rather than one | |
| Partial outcome | Purchases miss drinks brought from home, so compensation is invisible | Whole-day intake survey in a sample of schools as a secondary outcome | |
| Statistical conclusion validity | Few schools and clustered data make estimates imprecise | Analyze at the school level with appropriate standard errors; report confidence intervals | |
Note. Threat labels follow Shadish, Cook and Campbell; handling describes plans fixed before outcome data.
What This Design Can and Cannot Establish
If the parallel trends check holds and no major history event occurs, the design can estimate the program's effect on cafeteria sugary drink purchases in its first year with reasonable confidence. It cannot estimate effects on whole-day intake without the supplementary survey, cannot speak to weight, and cannot say whether first-year effects persist. Rossi et al. (2019) note that a design is credible to the extent its weaknesses are stated before the results are known; this memo is that statement.
Requests
I ask the program manager to confirm that installation dates for both groups will not change, to share the food service data agreements for all 70 schools by October, and to notify the evaluation team of any district beverage policy changes during the evaluation year.
References
Rossi, P. H., Lipsey, M. W., & Henry, G. T. (2019). Evaluation: A systematic approach (8th ed.). SAGE.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.
Wing, C., Simon, K., & Bello-Gomez, R. A. (2018). Designing difference in difference studies: Best practices for public health policy research. Annual Review of Public Health, 39, 453-469. https://doi.org/10.1146/annurev-publhealth-040617-013507
Reading the PHLT 8044 Week 6 instructions
The sixth assignment usually asks for a memo choosing one evaluation design and committing to it before data are collected. Faculty expect the memo to state the comparison, outcome, groups and time periods clearly, explain why alternatives were rejected and name the key assumption the design relies on and how it will be checked. The remaining threats to validity should be listed in the terms of a standard typology, each with a plan for handling or detecting it. Committing in advance matters because a design chosen after seeing data can be bent toward a desired result. Strong memos say what the design can and cannot establish. The memo is written for program decision-makers and an advisory group, so it should be clear and concise, typically three pages.
How the PHLT 8044 Week 6 example is put together
The memo opens with the decision: the comparison, the schools, the outcome and the periods, and why the timing of the commitment matters. A section explains why three alternatives were rejected, including randomization, which installation schedules ruled out. A section on the parallel trends assumption describes the pre-period check and what will happen if it fails. A six-row table lists remaining threats, history interacting with selection, instrumentation, diffusion, regression to the mean, the partial outcome and statistical conclusion validity, with how each could bias the estimate and how it will be handled. A paragraph identifies the most serious threat. The memo ends by marking the boundaries of what the design can show and by making three requests of the program manager.
PHLT 8044 Week 6 rubric: what earns full marks
The memo is graded on commitment, justification and candor about weaknesses. A memo that describes several designs without choosing one, or chooses one without naming its assumption and threats, misses the assignment. Credit goes to explaining why alternatives were rejected in terms of this program's constraints, to a concrete plan for checking the key assumption, and to threats named in standard terms with specific handling. Identifying the most serious remaining threat shows judgment. Graders also check that the memo does not overclaim: it should state what the design cannot establish. Clarity for a nontechnical program audience is part of the grade. Accurate citation of the validity typology and difference-in-differences guidance supports the technical content.
PHLT 8044 Week 6 help: mistakes that cost marks
Design memos falter when they present options instead of a decision. Choose one design, state it precisely and explain why the alternatives would not work for this program. Name the assumption your design depends on and how you will check it before outcomes are known. List the threats that remain in standard terms and say how you will detect or reduce each. Say what the design cannot establish. Ask the program for whatever the design needs from it. The same commitment format fits evaluations of any phased program.
Get a PHLT 8044 Week 6 example written to your instructions
Forward the Week 6 prompt and rubric and describe how your program rolled out, whether in waves, all at once or by eligibility rule, since that fact decides which comparisons exist. The memo returns in 24-48h, the first without charge. The sealant rollout and its waiting list are fiction unless you swap in a documented program.
More PHLT 8044 papers and PhD in Public Health samples
- PHLT 8044 Week 1: Monitoring Versus Evaluation Post
- PHLT 8044 Week 2: Water Program Evaluability Study
- PHLT 8044 Week 3: Water Access Program Logic Model
- PHLT 8044 Week 4: Water Program Evaluation Questions
- PHLT 8044 Week 5: Selection Bias Peer Reply
- PHLT 8044 Week 7: Indicators and Credibility Plan
- PHLT 8044 Week 8: Data Collection Schedule Plan
- PHLT 8044 Week 9: Preregistered Analysis Plan
- PHLT 8044 Week 10: First-Year Findings Memo
- PHLT 8044 Week 11: Second-Year Evaluation Plan
- PHLT 8042 Week 6: Quarantine Siracusa Critique
- PHLT 8040 Week 6: Septic Bill Clause Analysis
- PHLT 8034 Week 6: Landfill Link Peer Critique
- PHLT 8032 Week 6: Peer Replies on SPSS Output
PHLT 8044 Week 6 questions, answered
Where can I find a free PHLT 8044 Week 6 sample paper?
The full Week 6 design memo is on this page, committing a school water evaluation to a difference-in-differences comparison against waiting-list schools, with a table of remaining threats.
Why not randomize which schools receive stations first?
In this composite program, installation order was set by facilities schedules and principals' applications before the evaluation was planned, so randomization was not possible.
What happens if the parallel trends check fails?
The memo commits to revising the design and documenting the revision before outcome data are analyzed, rather than proceeding with an assumption the data contradict.
Why use two baseline years instead of one?
Two years show whether groups were on similar paths before installation and reduce the risk that a single unusual year makes program schools look different, a protection against regression to the mean.
Which threat is hardest to rule out?
History interacting with selection. Program and waiting-list schools are unevenly spread across districts, so a district-wide change in beverage contracts could affect one group more than the other.