Exercise: Descriptive Statistics 1

In this exercise, either on your own or in a group, you will carry out a small practical project that requires applying content from the introductory chapters on descriptive statistics.

If you would like to create an interactive version with direct input of your answers (for possible later discussion), you can complete this mini-course.

Introduction

Repeating this in an applied context is very important. These are the kinds of things that are always taken for granted—even though, after a certain point, you can no longer really “see” them. For example, when you read a scientific article, you simply have to assume that aspects such as measurement and sampling have been carried out properly—you cannot see this directly in the data or metrics presented in the article.

More specifically, this concerns the content of the following contributions:

• The relevance of statistics in psychology and work psychology
• The structure of the statistical data table
• Important terms
• Measurement and operationalization for descriptive statistics
• Data Collection for Quantitative Research
• Samples and Populations in Statistics

But let’s go through it step by step.

Step 1: Choose a hypothesis

We start our research design with the hypothesis (for the sake of simplicity, we will skip earlier steps such as defining the research objectives or research questions in this exercise).

Come up with a hypothesis or choose one of these example hypotheses. All subsequent steps should then relate to the hypothesis you have chosen:

1. A longer average sleep duration (over the past 7 days) is associated with greater concentration while studying.
2. More frequent use of spaced repetition per week is positively associated with the most recent quiz score.
3. Studying in a low-distraction environment leads to longer focused study time per day.
4. More Pomodoro sessions per day are associated with lower subjective procrastination.
5. Students who study in groups report higher learning motivation.

Output Step 1:

• One hypothesis.

Step 2: Determine the constructs

Determine which constructs occur in the hypothesis and what role they play in the hypothesis (in other words: what “type” of variable is involved here?). Also briefly research a definition of these constructs and briefly outline why it would make sense to test this hypothesis (relevance).

Output Step 2:

• A table naming and defining the constructs and describing their role in the hypothesis.
• A brief statement on the relevance of the hypothesis.

Step 3: Operationalization

Now it is about breaking down the respective constructs further and making them measurable. Here you can, if necessary, read again in the relevant article what needs to be considered when making constructs measurable (Measurement and Operationalization for Descriptive Statistics).

What we need are concrete questions that could be asked. You can obtain these through research (searching for existing scales) or by inventing your own items. Two or three items are perfectly sufficient. Note: This is of course only complete if you also define appropriate response formats.

Note: While you are at it, please also clarify which level of measurement (scale level) applies to each measurement of your constructs.

And because, of course, we want to do this properly from the outset, a codebook should also be created here. A simple table is sufficient.

Output Step 3:

• For each construct:
  • Items measuring the construct (including response format)
  • Clarification of the level of measurement for each item
• Codebook (across all constructs/items)

Step 4: Sampling

Create a hypothetical sampling plan for your (not-to-be-conducted) data collection. Define the population, the sample and, of course, the procedure that would be suitable for moving from the population to the sample (sampling procedure).#

Output Step 4:

• Description of the population
• Description of the sampling procedure
• Description of the sample

Step 5: Generate a dataset

Create a short example dataset that is consistent with Steps 3 and 4. Exactly how you proceed is up to you; in any case, the final result should be a table with 5–10 cases.

Output Step 5:

• Data table with at least 5 cases.

Step 6: Reporting / Reflection

Review steps 1–5 mentally once again and make notes on what was difficult and what was easy. If you like, you can also try to document the process in writing (something that is always required, at the latest, when doing academic research).

Output Step 6:

• Summary reflection
• Written report on the overall process, if applicable