Reason for Using Frequency and Percentage to Analyse Categorical Data

Background: Researchers conducting descriptive, survey-based studies must select data analysis techniques that are consistent with the level of measurement of the data collected. A recurring challenge in this process is uncertainty over whether the choice of frequency and percentage as descriptive statistics is influenced by the academic degree level of the study, and whether such techniques remain valid across both Likert-scale and non-Likert-scale data. Purpose: This article examines and justifies the use of frequency and percentage as the appropriate descriptive statistics for analysing categorical data regardless of the academic degree level of the researcher, and explains the rationale for applying a 50% benchmark when interpreting percentage distributions obtained from both Likert and non-Likert scale instruments. Methodology: The article adopts a conceptual, literature-based approach, synthesising established methodological sources on levels of measurement, descriptive statistics, and survey research practice. It draws on foundational works on scales of measurement, categorical data analysis, and Likert-scale interpretation, alongside guidance on survey research reporting, to build a coherent argument. Major Findings: The discussion establishes that the appropriateness of frequency and percentage is determined by the level of measurement of the data (nominal or ordinal) and the descriptive nature of the research design, not by the researcher's degree level. It further shows that both Likert-type and non-Likert items generate categorical response data at the item level, making frequency and percentage suitable for both. The 50% benchmark is found to be a mathematically grounded, transparent, and easily interpretable cut-off for distinguishing majority from minority response patterns, independent of scale type or study level. Conclusion: Frequency and percentage, supported by a 50% interpretive benchmark, are defensible and consistent descriptive statistics for categorical data across all degree levels and both Likert and non-Likert instruments, because their validity rests on the nature of the data rather than the qualification level of the researcher. Implications: The findings offer researchers, supervisors, and examiners a clear methodological justification for adopting frequency, percentage, and benchmark analysis in descriptive studies, reducing ambiguity in data analysis chapters and promoting consistency in the reporting and interpretation of categorical survey data across diploma, undergraduate, master's, and doctoral research.