Students learning together
Using Gorilla to teach sta­tis­tics and research methods

An age-old debate is whether we should teach sta­tis­tics not using manual cal­cu­la­tions, but also using soft­ware such as SPSS, MINITAB, JASP etc.

While many teach­ers and lec­tur­ers prefer manual cal­cu­la­tion, we believe there is a better way. We believe stu­dents gain the most valu­able skills by working with real data sets, pre-pro­cess­ing the data, and then running the rel­e­vant sta­tis­ti­cal tests as this pre­pares stu­dents to conduct their own research.

Cru­cial­ly, for exper­i­men­tal psy­chol­o­gists and behav­iour­al sci­en­tists, it’s not just about soft­ware for sta­tis­tics, but also the soft­ware for data pre-pro­cess­ing. The pre-pro­cess­ing is often the far bigger piece of work and often a more useful long term skill. In indus­try, far fewer roles require infer­en­tial sta­tis­tics. Many more roles benefit from data pro­cess­ing and data visu­al­i­sa­tion skills.

 

 

Tools:

Favoured tools are Excel (pivot tables & power pivot) and R, both of which are also gen­er­al­ly useful skills to learn. Stu­dents greatly appre­ci­ate it if the skills they learn are useful outside acad­e­mia — and Excel is ubiq­ui­tous. Favoured pack­ages for sta­tis­ti­cal analy­sis are JASP and SPSS.

If you’re new to Excel or R, we have some great resources on our support page to get you started.

Teach­ing Resources:

So, for me, best-in-class teach­ing takes stu­dents from the research ques­tion, to the data col­lec­tion, to the data pre-pro­cess­ing and then finally to the sta­tis­ti­cal analysis.

We have several FREE research methods lec­tures with data sets that can be used in class in Gorilla Academy. We cover topics like atten­tion, lan­guage, learn­ing & social influ­ence. But of course, the spe­cif­ic research topic is less impor­tant than the generic skills learnt.  Check out an example data analy­sis lecture from our selec­tive atten­tion study.

The generic skills learnt in these Gorilla Academy lec­tures are:

  • What raw behav­iour­al data looks like (one row per trial)
  • How to pre-process it (one row per participant)
  • How to select the appro­pri­ate sta­tis­ti­cal test for the data collected
  • How and when to use Excel, R, JASP

These lec­tures are perfect for stu­dents using Gorilla in their cap­stone or dis­ser­ta­tion projects, as these lec­tures will teach them many of the skills they need to be successful.

Jo Ever­shed

Jo is the CEO and co-founder of Caul­dron and Gorilla. Her mission is to provide behav­iour­al sci­en­tists with the tools needed to improve the scale and impact of the evi­dence-based inter­ven­tions that benefit society.