Descriptive statistics and sample profile table for your survey

What the two tables contain

• Sample profile table (participant characteristics): for each demographic question, the number and percentage of respondents in each category. When some respondents skipped a question, the table adds a valid percentage column based on those who answered, like the Valid Percent column in SPSS.
• Descriptive statistics table for each item and for each construct score: N, minimum, maximum, mean, standard deviation, skewness and kurtosis. Items that share a code (PU1, PU2, PU3) are averaged into a construct score computed on respondents who answered every item.

Worked example: the built-in sample data

Click “Use sample data” to load 220 student responses with four demographic questions and two constructs on a 5-point scale: Satisfaction (SAT1–SAT4) and Trust (TR1–TR3). The tool recognises the demographics on its own and produces:

• Sample profile: 132 women (60.0%), 87 men (39.5%) and 1 other (0.5%); most respondents were aged 20–22 (51.8%) and in their third year (35.5%). Three people skipped the income question, so income is reported with valid percentages: under VND 3m 38.7%, VND 3–5m 30.9%.
• Descriptive statistics: the Satisfaction score has M = 3.67, SD = 0.802 and the Trust score M = 3.38, SD = 0.843. At item level SAT4 has the highest mean (3.72) and TR3 the lowest (3.33).
• Normality: the largest absolute skewness is 0.309 and the largest absolute kurtosis 0.629, well inside ±2, so the data can be treated as approximately normal.

How to read the output

• Rating levels. With a 5-point Likert scale, many theses split the mean into five equal bands of (5 − 1) / 5 = 0.8: 1.00–1.80 very low, 1.81–2.60 low, 2.61–3.40 moderate, 3.41–4.20 high and 4.21–5.00 very high. A 7-point scale uses bands of 6/7 ≈ 0.86. This is a reporting convention, not a test, so check whether your department uses a different split. In the example, Satisfaction (3.67) is high and Trust (3.38) moderate.
• Normality. Skewness and kurtosis between −2 and +2 are commonly accepted as approximately normal (George & Mallery, 2010); some texts use a stricter ±1 for skewness. Values outside the range are highlighted in the table.
• Standard deviation. On a 5-point scale, an SD above 1 means opinions are quite spread out; an SD near 0 means nearly everyone chose the same option, which leaves little variance for correlation and regression.

How to report it in APA 7

Put the sample profile table at the start of the results chapter (or in the participants section of the method chapter, if your department prefers) and the descriptive statistics table before the reliability analysis. Report means and standard deviations to two decimal places and percentages to one, and give the number of valid responses in the table title, for example “Table 4.1. Sample characteristics (n = 220)”. A typical sentence reads: “The Satisfaction score had a mean of 3.67 (SD = 0.80), at the high level on the 5-point scale.” The tool's paragraph follows this pattern and cites George and Mallery (2010) for the normality range. Your results are also saved to the results chapter report builder, which places them before Cronbach's alpha and EFA so you can download the whole chapter as one Word file. The next step is usually Cronbach's alpha.

Frequently asked questions

Does it give the same numbers as SPSS?

Counts and percentages follow the SPSS Frequencies table, and skewness and kurtosis use the sample-adjusted formulas of SPSS and Excel SKEW and KURT, so the figures should match. Check one variable against SPSS the first time you use the tool.

Which columns are treated as demographics?

Columns with a small number of distinct answers, such as gender, age band or income band, or columns whose question names a demographic. You can tick or untick any column in step 2 and step 3 before building the tables.

What happens to blank answers?

In the profile table, blanks are shown as “(no answer)” and excluded from the valid percentage. A construct score is computed only for respondents who answered all of its items, so its N can be lower than the sample size.

Is my data uploaded anywhere?

No. The file is read and the tables are built in your browser; nothing is sent to a server.

References

• George, D., & Mallery, P. (2010). SPSS for Windows step by step: A simple guide and reference, 17.0 update (10th ed.). Allyn & Bacon.