Describe your sample and build Table 1

Every results chapter opens by describing the participants and the variables. Readers need to know who was studied and what the typical values look like before they can judge any test. Descriptive statistics also catch remaining data problems: a maximum of 199 for age or a mean of −20 for sleep is impossible to miss in a summary table.

Categorical variables: counts and percentages

Gender, faculty and counselling use are summarised with frequencies and percentages. Percentages should be based on the number of valid answers; if some are missing, say so.

The sample is 55.8% female. Faculties are roughly balanced, with Science the smallest group (20.4%). A third of the students (33.8%) have used the counselling service. In the text you would write: “The sample consisted of 240 students (134 female, 55.8%) aged 18 to 30 years.”

Continuous variables: centre and spread

For numeric variables, report a measure of the centre and a measure of spread. The mean and standard deviation (SD) are standard for roughly symmetric distributions. For skewed variables, the median and interquartile range (IQR) describe the typical case better, because a few extreme values pull the mean towards the tail.

Reading the numbers

Three things stand out. First, StudyHours is skewed: the mean (13.93) is almost two hours above the median (12), the maximum is 53 and skewness is 1.66. Most students study 8 to 17 hours a week (the IQR), and a few study far more. For this variable, report the median and IQR, and expect to need a non-parametric test or a robust check later. Second, stress and support are slightly above the scale midpoint of 3, with SDs under 1 point, which is typical for Likert scales. Third, final scores are on average 3.6 points higher than midterm scores; whether that difference is reliable is a question for the paired t-test in Lesson 6.

Skewness and kurtosis

Skewness measures asymmetry: zero for a symmetric distribution, positive when the right tail is longer. Kurtosis (excess kurtosis in the tool) measures how heavy the tails are compared with a normal distribution. Values between −2 and +2 are commonly taken as acceptable for parametric tests. StudyHours, with skewness 1.66 and kurtosis 4.41, sits at the edge for skewness and beyond it for kurtosis. The histogram shows why: a long tail of a few very hard-working students.

Descriptives by group

Choose a grouping variable in the descriptive statistics tool (for example Gender) to get the same table for each group. This is useful for checking group sizes before a t-test or ANOVA, and many journals expect descriptives by condition in experimental studies.

Building Table 1

Table 1 in most theses combines participant characteristics and the main study variables. Use APA style: a table number in bold, an italic title, horizontal lines only at the top, under the header and at the bottom, and a note explaining abbreviations. Report two decimals for means and SDs, and give the possible range of scales. You can paste any results table into the APA table formatter (Lesson 11) to get the correct layout for Word.

Table 1
Descriptive Statistics for the Study Variables (N = 240)
… rows as above …
Note. Stress and support were measured on 1–5 scales; higher scores indicate more stress and more support. Study hours were positively skewed (Mdn = 12, IQR = 9).