Test how missing follow-up changes the interpretation
See this course’s project requirements
A fictional program evaluation brief with denominator definitions, matched pre/post results, attrition sensitivity, project-rubric evidence, a noncausal conclusion and a next measurement plan.
Keep your working files in your own notebook or software. Save a private description, checklist, and work link in My Proof.
What to include
- Provide a stable dictionary for enrollment, completion, quiz and project measures.
- Retain all eight enrolled aliases and document matched/missing/exception records.
- Calculate denominator-specific rates and observed paired gain.
- Show explicitly hypothetical missing-score bounds and independent project-rubric evidence.
- Write a descriptive conclusion with causal limits and a next-evidence plan.
Review and revise
Use these criteria to check your work and choose what to improve.
- All rates and means use the stated cohort or matched subset.
- Missing and conflicting records are not silently dropped or invented.
- Quiz and reviewed-project evidence retain their different standards.
- The conclusion is descriptive and its limitations match the study design.
Want more background?
The idea in everyday life
Fictional learning program: eight enrolled aliases P1–P8 have baseline scores. Four post-test items score from 0 to 4; the exercise threshold is at least3. Completion means the supplied attendance/submission conditions were met, independently of quiz or project result. P1–P4 completed; their post scores are4,4,3,3 and pre scores2,3,1,2. P5–P8 did not complete and have missing post scores; their pre scores are2,1,0,1. A four-criterion project rubric was reviewed for P1–P4: P1/P3/P4 meet all criteria; P2 needs revision. No comparison group, random assignment or real participant data exists.
A sensitivity bound asks how an estimate could change under explicitly hypothetical missing values. It is not an imputation presented as actual evidence.
Test how missing follow-up changes the interpretation — supplied practice
A sensitivity bound asks how an estimate could change under explicitly hypothetical missing values. It is not an imputation presented as actual evidence.
Fictional learning program: eight enrolled aliases P1–P8 have baseline scores. Four post-test items score from 0 to 4; the exercise threshold is at least3. Completion means the supplied attendance/submission conditions were met, independently of quiz or project result. P1–P4 completed; their post scores are4,4,3,3 and pre scores2,3,1,2. P5–P8 did not complete and have missing post scores; their pre scores are2,1,0,1. A four-criterion project rubric was reviewed for P1–P4: P1/P3/P4 meet all criteria; P2 needs revision. No comparison group, random assignment or real participant data exists.
Fictional eight-learner evidence register
Paste at A1.
| AID | BPre score /4 | CPost score /4 | DCompleted program | EProject review |
|---|---|---|---|---|
| P1 | 2 | 4 | Yes | Pass |
| P2 | 3 | 4 | Yes | Revise |
| P3 | 1 | 3 | Yes | Pass |
| P4 | 2 | 3 | Yes | Pass |
| P5 | 2 | Missing | No | Not reviewed |
| P6 | 1 | Missing | No | Not reviewed |
| P7 | 0 | Missing | No | Not reviewed |
| P8 | 1 | Missing | No | Not reviewed |
Copy as plain text
1. Count unknown follow-up
List P5–P8 as the four missing post results and record the possible score range0–4.
You should see: Four unknown scores with total possible range0–16.
The instrument’s bounded scale makes a transparent scenario range possible.
2. Calculate the lower bound
For this hypothetical bound only, set all four missing scores to 0; divide observed total 14 by all eight.
You should see: All-cohort post mean lower bound1.75.
The hypothetical zeros are not entered as real learner records.
3. Calculate the upper bound
For this bound only, set all four missing scores to 4; add16 to 14 and divide by eight.
You should see: All-cohort post mean upper bound3.75.
The scenario uses the maximum allowed score, not a forecast.
4. Interpret against the cohort baseline
Use all-eight baseline mean 12/8=1.5; state hypothetical cohort-change range+0.25 to+2.25 points and keep observed paired gain separate.
You should see: The range describes assumptions; the observed paired gain remains1.5 for four learners.
Bounds do not remove selection bias, measurement limits or the lack of a causal comparison.
What to finish
A labeled missing-follow-up sensitivity table and limitation note.
Watch for this
Reporting hypothetical minimum or maximum scores as if missing learners actually took the test.
How can I tell I understand it?
Sensitivity, observed paired change and causal interpretation remain separate.
Quick knowledge check
Answer all 3 questions. Get at least 2 right and the lesson completes automatically.
You’ll get an explanation for each answer. Signed in? Your results save to your account.
Pass the check. We’ll save the progress.
Answer all 3 questions above and get at least 2 right. You can review the explanations and try again.
Go to the knowledge checkTools, templates, and lesson sources
Tools and templates
Tools open in a new tab so you can keep your place here.
Sources and further reading
Explore the references behind this lesson.
Source & access notes · checked 2026-10-01
This official reference supports the mapped concepts. The supplied exercise inputs and operating rules are original fictional practice, not universal legal requirements or qualification evidence.
Access: official search extract of program-evaluation toolkit; fullPDFreadererror on 2026-10-01; external resources are optional.
Reuse: Original fictional teaching and practice; source concepts paraphrased and linked, with no copied provider exercises.
Source & access notes · checked 2026-10-01
This official reference supports the mapped concepts. The supplied exercise inputs and operating rules are original fictional practice, not universal legal requirements or qualification evidence.
Access: official IES missing-data report metadata/search extract; fullreadererror on 2026-10-01; external resources are optional.
Reuse: Original fictional teaching and practice; source concepts paraphrased and linked, with no copied provider exercises.
Report a problem with this lesson
Tell us what went wrong. Please leave out personal information and private project contents.