Stats Assistance
Defense · one doctoral project, after data collection

The output is not
the chapter.

It is the part of Chapter 4 you can defend line by line.

Open your data file. Defense runs the five tests a quantitative results chapter needs, in the order a committee reads them: instrument, descriptives, means, correlation, discussion. You get APA 7 tables, the reason each test was chosen over its alternative, and a Word download. Every number is computed from your file. No language model writes a sentence.

Run the demo file no signup
Your file never leaves this browser APA 7 tables + Word download $199 one project · $49 instrument check

1 Instrument · reliability of the rating items Demonstration data

Loading the demo file…

The sequence is the chapter

Five steps, and a gate at every one.

Order is not a style preference. A weak measure invalidates every test after it, and a variable with no spread has nothing to compare. Defense stops where a statistician would stop, and tells you why. Pick a step to see what it hands you.

Every table here is calculated from the demo file: 250 respondents, six Self-Awareness items, Age Group, Gender, and a job performance score.

Loading the demo file…

Read all five steps as one report Then open your own file. It is read and calculated in this browser.
How the prose is made

“Not written by AI” is checkable. Here is the mechanism.

A sentence template was written once, by a person, and it does not change between projects. Your file fills the slots. That is the whole trick, and it is why the output reads the same way for every student and cannot invent a citation, a number, or a conclusion.

The template, fixed, written by a human

The demo file, dropped into the slots

Calculated when the demo file loads.

Calculated from your file

  • Cronbach's alpha and item-total correlations
  • Means, SDs, ranges, group counts
  • Welch t or ANOVA on the groups that survived
  • Post hoc comparisons when there are three or more groups
  • Effect sizes
  • Pearson correlation, kept apart from the group tests

Never generated

  • Your literature review, hypotheses, or citations
  • A theoretical explanation for why a result came out that way
  • Recommendations for practice or policy
  • Any sentence that was not written in advance by a person
  • A conclusion your design cannot support
  • Your Chapter 4. The output is the evidence you write it from
Price

One project, one price.

A statistician doing the same five steps quotes $400 to $2,700. Defense is not that service and does not replace it. It is the sequence, run on your file, today.

This site is new and has no testimonials yet. Rather than ask you to trust a claim, the demo file runs the whole packet for free. Read the output first and decide from that. Open your own file when you are ready: $199 for the full path, $49 for the instrument check. The file is read in this browser and never uploaded. Close the tab and the file is gone. After you pay, a sign-in link finds you by email. Open the same spreadsheet again and the chapter runs.

$49
Instrument check

Step 1 only. A verdict on whether the measure can carry the claim.

  • Alpha and item statistics
  • Flagged items
  • A plain keep-or-stop verdict
  • Best used before you finish collecting
Before you buy

The questions people actually ask.

Answered directly, including the uncomfortable one.

Is using this allowed? Will my committee have a problem with it?

Ask your chair, and ask before you buy. Programs differ. What you would be disclosing is that a tool calculated the tests and formatted the tables, which is what SPSS, R, and a hired statistician also do. What is not happening is a model writing your argument. If your program requires you to name your analysis software, name Defense the way you would name SPSS.

How do I know a language model did not write it?

Run the demo file twice, or run two similar files. The sentences come back identical apart from the numbers, because they are templates with slots, shown above. Generated prose varies every time. This does not. That is the test, and you can run it yourself before paying.

Where does my data go? I have IRB obligations.

Nowhere. The file is read and calculated inside this browser tab, and never uploaded to a server. Close the tab and the file is gone. After you pay, sign-in is a link to your email, not a stored copy of the data. You open the same spreadsheet again. If your protocol requires that participant data not leave your device, that condition is met, and it is worth putting that sentence in front of your IRB rather than taking ours for it.

What does my file need to look like?

CSV or Excel. One row per respondent, one column per question, and a header row with names you recognise. Rating items as numbers, groups as labels or codes. Reverse-scored items should be recoded before you open the file, so the reliability step reads the scale the way you intend it.

Why Welch instead of the t-test I was taught?

Student's t assumes the two groups have equal variances. Dissertation samples rarely do, and unequal group sizes make the violation worse. Welch does not make that assumption and loses almost nothing when variances happen to be equal, so it is the safer default. The report states the choice, so you can defend it.

When should I hire a human statistician instead?

When your design is outside these five steps: regression with several predictors, mediation or moderation, factor analysis, multilevel or repeated-measures data, or serious missingness. Also when a committee has already rejected an analysis and you need someone to defend a rebuild with you. Consulting is that service, and it is still the right answer for those cases.

See it run on the demo file before you spend anything.

The whole packet, all five steps, the tables and the chair lines. Then decide whether your own file is worth $199.

Run the demo file

Free calculators stay free. Defense is the sequence, not a calculator.