If you haven’t found a dataset to use for the final project you can find one in the folder here. There are 10 different datasets contained in that folder, numbered 0 to 9. You should pick the dataset that has the same number as the last digit of your matriculation number. The datasets are slightly different so you may draw different conclusions depending upon which dataset you use. All datsets are tab-deliminted and include column headings.
In this (simulated/fake) study a group of researchers where interested in increasing the cognitive performance of students. To do this they recruited 300 students to take part in a 3 month study. During the study participants received several interventions- based on diet, drugs and exercise- that the researchers thought may effect their cognitive performance.
For the diet intervention participants were randomly assigned to either a plain diet condition (coded 0) where they ate their normal diet, or a superfood diet condition (coded 1) where they ate a special diet heavy in “superfoods” like kale, quinoa etc. For the drug intervention participants either: 1) received no drugs; 2) were told that they were taking a special cognitive-ehancement drug, but actually received a placebo; or 3) received an experimental cognitive-enhancement drug called factor-x. Finally, participant could also choose to take part in either a weightlifting class or a yoga class.
At the end of the 3 months participants had to complete several cognitive tasks including: an IQ test; a General Knowledge test (which had 50 questions each worth 1 point); an n-back task (this is a memory based task often ued in psychology studies); and a stop signal task (another psychology task which measures impulsivity/response inhibition). For both the n-back and stop signal task performance is measured as the percentage of trials where the participant was correct; with high scores indicating good memory performance and good impulse control respectively.
In addition to the above measures, gender and age were also recorded. The data in the datafiles is the “raw” data, where the researchers have not yet checked for coding errors in any of the measures (e.g. negative IQs, percentages greater than 100 etc.).