Data collection is a foundational step in any statistical analysis, enabling professionals to gather relevant information for decision-making across various sectors. Whether in healthcare, agriculture, business, or public administration, accurate data collection ensures reliable insights and effective planning. This chapter introduces the principles of data collection and explores primary and secondary methods, equipping students with practical skills to gather and interpret data in real-world Kenyan contexts.

4.1 Introduction to data collection

Data collection involves systematically gathering information to address specific questions or problems. In Kenyan workplaces such as county referral hospitals, banks, or agricultural cooperatives, data collection supports monitoring, evaluation, and strategic planning. The process requires careful attention to accuracy, relevance, and ethical considerations. It forms the basis for statistical analysis, quality control, and evidence-based decision-making.

4.1.1 Purpose of data collection

Data collection serves multiple purposes in professional settings:

• Inform decision-making: Accurate data enables managers to make informed choices based on factual evidence rather than assumptions.
• Monitor performance: Organizations track progress toward goals, such as patient recovery rates in hospitals or loan repayment rates in SACCOs.
• Identify trends: Data helps reveal patterns over time, such as seasonal crop yields for farmers or customer footfall in retail outlets.
• Support research: Data underpins studies aimed at improving processes, products, or services.
• Ensure accountability: Transparent data collection allows stakeholders, including funders and regulators, to assess organizational effectiveness.

4.1.2 Types of data

Data collected can be classified into several types relevant for analysis:

• Quantitative data: Numerical measurements such as income, temperature, or attendance numbers.
• Qualitative data: Non-numerical information like opinions, descriptions, or observations.
• Continuous data: Data that can take any value within a range, for example, blood pressure readings.
• Discrete data: Countable values, such as the number of employees or vehicles.
• Categorical data: Data sorted into categories, for example, types of crops or customer satisfaction levels.

4.1.3 Data collection challenges

Effective data collection faces various challenges in Kenyan workplaces:

• Incomplete data: Missing values due to non-response or oversight can bias results.
• Inaccurate data: Errors in measurement or recording reduce reliability.
• Cost constraints: Limited budgets may restrict the scope or frequency of data gathering.
• Respondent bias: Participants may provide socially desirable answers rather than truthful ones.
• Data privacy: Ensuring confidentiality and ethical use of personal information is critical.

4.1.4 Steps in data collection

Data collection follows a systematic process to ensure quality and relevance:

1. Define objectives: Clarify what information is needed and why.
2. Select data types: Decide on quantitative or qualitative data based on objectives.
3. Choose collection methods: Determine whether primary or secondary data sources will be used.
4. Design instruments: Develop tools such as questionnaires or checklists.
5. Collect data: Implement the collection plan in the field or office.
6. Validate data: Check for completeness and accuracy before analysis.

Chapter Summary

This chapter explored the fundamental concepts of data collection, emphasizing the importance of gathering accurate information for statistical analysis. It distinguished between primary data, which is collected firsthand, and secondary data, which is obtained from existing sources. Various sampling techniques were discussed, including probability methods that ensure every member of a population has a known chance of selection, and non-probability methods where samples are chosen based on specific criteria. The chapter then examined different ways to present data effectively through tables and diagrams, such as frequency distribution tables, bar charts, pie charts, histograms, and frequency polygons. It also covered types of graphs used in statistics, including basic time series graphs, z-charts, Lorenz curves, and semi-logarithmic graphs, each serving distinct analytical purposes. Finally, the use of cumulative frequency curves, known as ogives, was explained as a tool for understanding the distribution of data cumulatively. Together, these topics provide a comprehensive foundation for carrying out elementary statistics in various practical contexts.