Questionnaire Item-Level Descriptive Statistics Analyzer

Scope

This calculator describes participant responses for numerical questionnaire items and summarizes participant-level mean item and total scale scores among complete cases. It does not calculate reliability coefficients, item–total correlations, factor analysis, reverse-scored totals, imputed values, or pass–fail decisions about item quality.

Formulae

Response percentages use the non-missing responses for each item as the denominator. Missing percentages use all imported participants. The mean is the arithmetic average and SD is the sample standard deviation with denominator n − 1. Skewness is the adjusted Fisher–Pearson sample coefficient, equivalent to Excel SKEW. Kurtosis is the bias-corrected excess-kurtosis estimator, equivalent to Excel KURT.

Assumptions and limitations

• Rows represent participants and item columns contain numerical response codes.
• Means, SDs, skewness, and kurtosis treat response codes as numerical scores; their usefulness depends on the response scale and intended interpretation.
• Missing responses are excluded item by item. The overall scale summary includes only participants with complete responses across all items; no values are imputed or prorated.
• The overall summary assumes that all items belong to the same intended scale, use compatible response coding, and have already been reverse-scored where necessary.
• Skewness requires at least three valid responses and non-zero variance. Excess kurtosis requires at least four valid responses and non-zero variance.
• Distributional statistics and generic thresholds do not by themselves determine whether an item should be retained or removed.

Rounding

Counts are integers, response and missing percentages use one decimal, and mean, SD, skewness, and excess kurtosis use three decimals. Calculations retain full internal precision.

Validation

Version 1.0.0 was tested against hand calculations and Excel-compatible definitions for response counts, missingness, sample SD, adjusted skewness, and bias-corrected excess kurtosis. The Excel workbook is reopened during automated tests to confirm worksheet structure and numeric values. Availability supports independent comparison and is not a regulatory validation claim.

Excel export

The workbook contains a formatted reporting table, an overall scale summary, machine-readable numeric results, and analysis notes. Respondent-level source data are not included.

Privacy

All parsing, calculation, and workbook generation occur in browser memory. The application does not upload, transmit, or retain participant IDs, item headings, responses, or results. Do not paste direct identifiers or confidential information.

References

1. Joanes DN, Gill CA. Comparing measures of sample skewness and kurtosis. J R Stat Soc Series D. 1998;47:183–189.
2. DeVellis RF, Thorpe CT. Scale Development: Theory and Applications. 5th ed. Sage; 2021.
3. Streiner DL, Norman GR, Cairney J. Health Measurement Scales: A Practical Guide to Their Development and Use. 5th ed. Oxford University Press; 2015.
4. Microsoft. SKEW function. Microsoft Support.
5. Microsoft. KURT function. Microsoft Support.