For the students with no or little background in Math, this is for you. Data, broadly, can be categorized as either primary or secondary, yet many universities in Kenya predominantly expect students to rely on primary data. This conservative stance on data sources can limit the objectivity and flexibility of research. Allowing the use of secondary data could provide students with richer, verifiable information while easing the logistical and financial burdens of extensive fieldwork. While primary data has its merits, the potential benefits of secondary data are often underexplored in Kenyan academia—a topic worthy of deeper discussion another day.

The immediate challenge arises when students embark on collecting primary data, particularly when designing questionnaires or interview schedules. To do this effectively, students must first understand the type of data they are dealing with, as this determines how the data can be analyzed and interpreted.

Types of Data

Broadly, data is classified into quantitative and qualitative types.

Quantitative Data [Numerical Data]

Quantitative data refers to information that can be measured numerically and is suitable for statistical analysis. It answers the “how many” or “how much” questions. Quantitative data can be further classified into:

1. Discrete Data – This type of data can take on only specific, distinct values. They are countable and cannot be meaningfully subdivided. Data like number of students in a class, number of books in a library, number of children in a household.

Discrete data often has a true zero, representing a complete absence of the measured attribute. For instance, zero books literally means no books.

1. Continuous Data – Continuous data can take on any value within a given range and can be subdivided infinitely. For example, Height of students, temperature readings, income in Kenyan shillings.

Continuous data may have a zero point, but it does not always represent the absence of a quantity in the same way as discrete data. For example, a temperature of 0°C does not mean there is “no temperature.”

Choosing the right quantitative questions is critical. Misclassification can lead to invalid results; for example, asking for income ranges but treating them as discrete counts may produce misleading averages or percentages.

Qualitative Data [Categorial Data]

Qualitative data, on the other hand, refers to descriptive information that cannot be measured numerically in a meaningful way but can be categorized or ordered. Qualitative data is often subdivided into:

1. Nominal Data – Categories that have no inherent order. Examples of such data are aspects like Gender (male/female), type of mobile phone (Android/iOS), marital status (single/married/divorced).

Nominal data does not have a true zero, since the zero point is meaningless in this context.

1. Ordinal Data – Categories that have a logical order [from the word order] but where the intervals between categories are not necessarily equal. Examples of such data include; Education level (primary/secondary/tertiary), satisfaction ratings (unsatisfied, neutral, satisfied), class rank.
2. Characteristic: Ordinal data may or may not have a true zero depending on context, but the main focus is on relative ranking rather than absolute measurement.

The design of questions for qualitative data is just as critical as for quantitative data. Poorly designed ordinal questions, for instance, can make it difficult to interpret trends or perform meaningful statistical analysis.

Importance of Matching Question Types to Data Types

The type of questions you ask must align with the data type you intend to collect because it directly affects:

• The accuracy of your analysis
• The reliability of conclusions
• The choice of statistical tools you can legitimately use

For instance, trying to calculate a mean for nominal data is meaningless, while using frequency counts for ordinal or nominal data is entirely appropriate. Understanding these distinctions ensures that students’ research outputs are robust and credible.

The Role of SPSS and Other Analytical Tools

Software such as SPSS (Statistical Package for the Social Sciences) has been the standard tool for quantitative data analysis in Kenya. SPSS allows researchers to efficiently manage large datasets, perform descriptive and inferential statistics, and visualize findings. Key reasons SPSS is critical include:

• Ease of handling large datasets
• Automated statistical testing
• Versatility in generating graphs and tables

However, with the evolving landscape of data science, Kenyan lecturers and universities should consider encouraging the use of alternative tools such as R, Python, Stata, or even Excel for advanced analytics. These tools provide more flexibility, advanced modeling capabilities, and integration with modern data sources. Accepting a broader array of tools would empower students to explore new methodologies and produce richer, more objective research findings.

Conclusion

In summary, one of the significant challenges facing Kenyan postgraduate students is not just the collection of data but understanding what type of data to collect and how to analyze it. Whether dealing with primary or secondary data, quantitative or qualitative, careful consideration of data types, question design, and analytical tools is critical to producing credible, verifiable research. By expanding the acceptance of secondary data and diverse analytical software, Kenyan universities could foster a more objective, innovative, and student-friendly research environment.

 For your Thesis and Dissertation Writing Services

Okite Linus

+254703437944