Scale scoring and item analysis
Audit item direction against your codebook, reverse-score, and compute documented scale scores.
See the workflowFREE CHECKLIST · PDF
Audit your data
before you doubt
your hypothesis.
A coding or scoring error can hide a real effect and produce a convincing null result. This two-page checklist walks through the checks to run first: missing codes, reverse-scoring, merges, scale scores, and a test on a relationship you know should exist.
Free · Two pages · Delivered once by email, with no mailing list.
WHAT'S INSIDE
Six sets of checks,
in a sensible order.
Each item is a box you can tick.
Work through them before you change your plan.
- 1Set upProtect the original file, open the codebook, and start a change log.
- 2Check each variableRanges, missing-value codes, frequencies, data types, and category direction.
- 3Check scales and scoresItem direction, reverse-scoring, recomputed scores, the missing-item rule, and reliability.
- 4Check rows and structureRow counts at every merge, filter, and reshape, plus IDs and exclusions.
- 5Check that the data behaveTest a known relationship, compare descriptives with norms, and plot.
- 6Close the loopFix the data, rerun the planned analysis, and document what you found.
WHY IT'S WORTH THE TIME
In the Analysis Clinic's worked example, skipping a single reverse-scoring step shrinks a clear effect (an estimate of about 3.0) to an apparent null (about 0.1). The data and the hypothesis never changed. The preparation did.
The example uses constructed teaching data, not a real study, and shows one way the error can happen. Read the full Case and run the R or Python code
AFTER THE CHECKLIST
When you want
more than a checklist.
Use these once you know where the problem is.
Data cleaning and audit
Turn a raw dataset into a documented analysis-ready file, with a row-level trail of every change.
See the workflowData preparation review
You prepare the data. A consultation helps you check it step by step so you can explain every decision.
How support is scopedA FEW THINGS YOU MIGHT BE WONDERING
Good questions.
Honest answers.
Is the checklist really free?
Yes. Enter your email and the two-page PDF is sent to you once. There is nothing to buy.
What happens to my email address?
We use it to send you the checklist once and to send a notice to our inbox that the checklist was requested. You are not added to a mailing list, and we do not sell or share your address.
Will I get a newsletter or follow-up emails?
No. You receive one email with the checklist attached. We only write again if you reply or book a consultation.
Who is the checklist for?
Doctoral and master's students whose analysis returned a surprising or null result and who want to rule out a data preparation error before changing their hypothesis or analysis plan.
Does the checklist replace my advisor's guidance?
No. It is educational. It helps you find common preparation errors, but it does not certify that a dataset is correct or replace your program's requirements.
STILL STUCK?
Bring the result
that doesn't add up.
Meet with Dr. Yel for a free 30-minute consultation about your data, your analysis, or your committee's feedback.
Book a free consultationCan't find a time that works? Email bookingrequest@dissertationstatshelper.com with a few times you're available and your time zone, and we'll find a time. It's a new address, so if you don't see our reply, please check your spam folder.