What thesis statistics help is a student allowed to get?

Direct answer: Universities that run their own statistics services describe the help as advice, not as analysis done for you. Oregon State University (2020) rules out data entry and analyses by its consultants, and Missouri State University (n.d.) says its consultants advise and guide rather than design a student's thesis project.

The most reliable way to learn where the line sits is to read how universities describe their own services. Oregon State University (2020) answers the question of how a student gets help with statistics for a proposal by stating that "the assistance given is primarily advice" and that "consultants will not do data entry or statistical analyses for you". Missouri State University's thesis approval form for its R Stats consultants says they intend "to provide statistical advice and guidance and not to design and analyze a student's thesis or research projects data" (Missouri State University, n.d.). The same form asks the student to obtain the permission of the committee before meeting a consultant independently, and to tell the committee the general content of the meeting (Missouri State University, n.d.).

Two things follow from those pages. First, advice is the default offer: someone explains how to choose a method, how to run it and how to read it, and you carry out the analysis yourself. Second, your supervisor is part of the arrangement, not an obstacle to it. Telling your supervisor who is helping you with the statistics protects you if anyone later asks how the analysis was produced.

These two services are American, so they are examples of a pattern rather than your university's rule. UK and Australian universities publish their own statistics, maths and academic skills support, and your handbook or library page will tell you what yours offers.

Direct answer: You can ask for help choosing the test, checking its assumptions, learning the software, running the analysis with you watching, and understanding the output. Field (2018) treats assumptions such as normality, homogeneity of variance and independence as things the analyst must check before trusting a parametric test, and a helper can show you how.

Think of the statistics in a thesis as six jobs, and note which ones a helper can share and which one is yours alone.

1. Choosing the test. The test has to match the research question, the type of each variable and the study design. A helper can walk you through that logic, and our guide on how to choose the right statistical test shows the questions to answer first. The decision still has to be one you can defend to your examiner.
2. Checking assumptions. Field (2018) covers the assumptions behind common tests, including normality, homogeneity of variance and independence, and how to look for outliers. A helper can show you which plots and tests to run and what a violation means for your choice of method.
3. Operating the software. SPSS, R, AMOS and SmartPLS each have a learning curve unrelated to your research skill. Having someone sit with you while you run the commands is closer to tuition than to ghost-work, especially when you keep the syntax file and can rerun it yourself.
4. Running the analysis. This is the grey zone. Running a test together, with you at the keyboard or with the syntax handed over and explained, is advice. Receiving a finished results table you cannot reproduce or explain is the point where many services and many universities become uncomfortable.
5. Interpreting the output. A helper can explain what a coefficient, a confidence interval or a model fit index means. What your results mean for your research question, and how they connect to the literature, is the argument of your thesis, and that has to be yours.
6. Reporting. The American Psychological Association (2020) sets the conventions for reporting statistics in APA style. A helper can check that your tables and notation follow them, but the sentences that report and discuss your findings should be in your own words.

If you can explain every step of items 1 to 5 to your supervisor without notes, you are on safe ground. Four sources sit behind this guide: Field (2018, 5th ed., Sage) and APA (2020, 7th ed.) for the 2 books, plus Oregon State University (2020) and Missouri State University (n.d.) for the 2 services. If you cannot, the help has gone further than advice, whatever it was called.

Direct answer: The research question, the decision about which method answers it, the interpretation and the writing must remain yours. Both the Oregon State University (2020) and Missouri State University (n.d.) pages describe advice rather than analysis done on your behalf, and the thesis is assessed as your demonstration that you can do the work.

A thesis is not only a set of results. It is evidence that you can pose a question, choose a method to answer it and judge what the answer means. That is why universities hesitate over any arrangement in which someone else produces the analysis. The test you picked, the assumptions you checked and the reasons you accepted or rejected an outlier are all things an examiner may ask about at a viva or defence.

A practical test is reproducibility. Can you reopen the data file and the syntax, rerun the analysis and get the same tables? Can you say in two sentences why this test and not another? If the answer to both is yes, the help has built your capability. If either answer is no, the help has replaced it. Our guide on dissertation coaching versus going it alone discusses the same distinction for the thesis as a whole.

Disclosure is part of the picture. Many universities expect students to acknowledge support in the preface or declaration, and Missouri State University (n.d.) asks for the committee's permission before independent consultation. Check what your own handbook says, and when in doubt ask your supervisor before you start rather than after you submit.

Direct answer: Choose a helper who teaches and hands over the files you need to reproduce the work, who matches your discipline and software, and who agrees in advance that the interpretation and the writing remain yours. Be cautious of anyone who promises particular results, which would be a sign they are not following the data.

Before you commit, ask these questions and notice how clearly they are answered.

• Will I be able to rerun the analysis myself? A good answer includes the cleaned data file, the syntax or script and a short note on every decision made along the way.
• Do you work in my software and my kind of study? A helper experienced in survey research with SPSS is not automatically the right person for structural equation modelling in AMOS or partial least squares in SmartPLS.
• Who writes the results and discussion? The answer should be you. A helper may read and comment on your draft, but should not draft it.
• What happens if the data do not support my hypothesis? A trustworthy helper says the finding is the finding. Anyone who implies they can produce a significant result is a red flag, because the analysis then follows the wish rather than the data.
• Is my supervisor aware? A helper who encourages you to keep the arrangement from your supervisor is asking you to take a risk for them.

A simple checklist covers it: 5 questions asked, 3 files received (the cleaned data, the syntax and a log of decisions), and 1 conversation with your supervisor before the first session. Also ask for a sample of the explanation they give, not only a sample of the output. How they teach is the best evidence of whether the help will leave you more capable.

Direct answer: A MAAS mentor with the right discipline agrees the analysis plan with you first, runs the analysis with you, and explains what each step does. You receive the cleaned data, the syntax and the tables so that you can repeat the analysis, and you write the results and discussion yourself.

MAAS's data analysis support for SPSS, R and Python is built around that boundary. The mentor acts as advisor and tutor, you remain the author of the thesis, and nothing is submitted on your behalf. If you are not sure which of the six jobs above you need, a free 20-minute consultation with MAAS, a Ho Chi Minh City company, is a conversation about where you are with your analysis, with no obligation. You can book one here.