Statistics and research methods make me anxious, and my project needs them
Understand where statistics anxiety comes from, and build enough confidence to plan and run the analysis your project needs.
In short
Start here, today. Write three sentences: your research question, the kind of data you have or will collect (for example counts, categories, scores or text), and how the data were or will be collected. Take those sentences to your next supervisor meeting or a statistics consultation. This takes about 15 minutes.
After that, work through these steps. Choose one to jump to it.
Step by step
Tick each step as you finish it. Open a step for the how.
0 of 6 steps done
Name exactly what you don't understand
“I'm bad at stats” is too big to act on. “I don't know whether to use a t-test or a regression for my two groups” is a question someone can answer. Keep a list of specific questions as they come up and work through them one at a time.
Start from your question, not the software
Before you open a program, sketch your design on paper: what you're comparing or predicting, what kinds of variables you have and whether the same people were measured more than once. That sketch tells you most of what you need to choose an analysis. It also makes conversations with experts far more productive.
Use the help that exists
Many universities have a statistical consulting service, a data or methods librarian, or drop-in help run by graduate students. Simon Fraser University's psychology consulting service, for example, asks students to bring a written summary of their research and a diagram or table of the study design, and a first consultation usually runs 30 to 60 minutes. Consultants help you reason through the analysis, but the decisions and the work stay yours.
Learn one method well
Pick the method your project needs and learn it properly from one good source, such as a textbook chapter or a course recording. Practise on a small example dataset before using your own data. Write down each step you take, so you can repeat it and explain it in your methods section.
Record your analysis decisions
Keep a short entry in the Decision Log for each choice: which variables you included, how you handled missing data, which test you ran and why. Your supervisor and examiners will ask about these choices, and the log turns a nervous moment into a straightforward answer. It also saves you from forgetting what you did three months ago.
Ask the basic question
If you don't understand something in a meeting, say so: “Could you explain what that model is doing, in plain terms?” Most supervisors would much rather explain now than read a methods chapter built on a misunderstanding. Put your questions in Supervisor Meeting Prep beforehand so they don't slip away in the moment.
Words you can use
Is this you? Signs to look for
- You chose your topic partly to avoid numbers, and now your supervisor wants a quantitative part.
- You open the software, look at the output and close it again.
- You nod along in meetings when people mention models you don't understand.
- You'd rather struggle alone for weeks than ask a question that might sound basic.
Why this happens
Statistics anxiety is common and well studied. In their review of the research, Onwuegbuzie and Wilson describe how many students report high levels of anxiety in statistics and research methods courses, and how that anxiety can affect performance in its own right. Researchers treat it as having several parts rather than one, including worry about interpreting results.
One of those parts matters especially in a first research project. Chew and Dillon, drawing on earlier work by Cruise, Cash and Bolton, list fear of asking for help as one of the six commonly measured dimensions of statistics anxiety. It's what keeps students stuck for weeks on a problem that a consultant could clear up in half an hour. Avoiding the problem makes the anxiety bigger, and the anxiety makes the problem look bigger.
Most master's projects need a small number of methods done carefully, not advanced mathematics. The right analysis follows from your question, the kind of data you have and how the study is designed. Once you can say those three things clearly, the choice usually narrows to a few options you can learn.
Worth a look (6)
- “Statistics anxiety: nature, etiology, antecedents, effects, and treatments”, Onwuegbuzie and Wilson, Teaching in Higher Education (2003)A review of research on what statistics anxiety is, where it comes from, how it affects students and what helps.
- “Statistics anxiety and the Big Five personality factors”, Chew and Dillon, Procedia: Social and Behavioral Sciences (2014)A short study that sets out the six commonly measured dimensions of statistics anxiety, including fear of asking for help.
- Statistical consulting, Simon Fraser University Department of PsychologyAn example of what a university statistics consulting service offers graduate students and how to prepare.
- Discovering Statistics Using IBM SPSS Statistics, by Andy FieldA widely used, informal textbook that explains common statistical methods step by step. Versions also exist for R.
- Sage Research MethodsBooks, case studies, videos and practice data sets on research methods, from interviews to regression. Your library usually provides access.
- PRISMAThe reporting guideline for systematic reviews and meta-analyses, with the checklist, flow diagram templates and extensions for scoping and other reviews. Many journals expect you to follow it.
When to bring someone in
Bring your supervisor in early, ideally before you collect data, so the analysis plan fits the design. Your university's statistical consulting service, research methods librarian or graduate methods workshops can help with specific questions, and it's good practice to tell your supervisor what help you've had. If anxiety about statistics or your project is affecting your sleep, mood or ability to work, talk to your counselling service.
Find the right person, or talk to Ify