Unit 4: Types of Data in Research - Subjective Questions

Primary data refers to original information collected directly by the researcher for a specific research problem or objective. It is collected firsthand and is designed to meet the exact requirements of the study.

Major sources and methods of collecting primary data include:

• Observation: Recording the behavior, actions, or conditions of participants or events.
• Interviews: Collecting information through face-to-face, telephone, or online interaction with respondents.
• Questionnaires: Using a structured set of written or digital questions to obtain responses.
• Schedules: Questions are asked by an enumerator who records the respondent's answers.
• Experiments: Data are collected by manipulating one or more variables and observing the effects.
• Focus groups: A moderated group discussion used to explore opinions, attitudes, and experiences.

Primary data are generally more relevant and current, but their collection may require considerable time, money, and planning.

Primary data are collected directly by the researcher for a particular investigation.

Advantages:

• They are closely related to the objectives of the study.
• They are usually current and reflect present conditions.
• The researcher has greater control over the method, sample, and quality of collection.
• The data can be collected in the required form and level of detail.
• Confidential or specialized information can be obtained when appropriate safeguards are used.

Limitations:

• Collection is often expensive and time-consuming.
• Trained investigators may be required.
• Non-response, interviewer bias, and respondent bias can affect accuracy.
• Designing instruments and selecting a suitable sample require expertise.
• Large-scale studies may be difficult to organize.

Therefore, primary data are highly useful when existing information is inadequate, but their cost and operational difficulties must be considered.

Secondary data are data that have already been collected, processed, and recorded by another person, organization, or researcher for a purpose that may differ from the current study.

Important sources include:

• Government publications: Census reports, economic surveys, health statistics, labor reports, and statistical abstracts.
• International organizations: Reports and databases published by organizations such as the United Nations, World Bank, and World Health Organization.
• Research institutions: University studies, working papers, research reports, and dissertations.
• Commercial sources: Market research reports, business databases, industry publications, and company records.
• Academic sources: Books, journals, review articles, and conference proceedings.
• Internal organizational records: Sales records, financial statements, employee records, and customer databases.
• Electronic sources: Online repositories, digital libraries, official websites, and open-data portals.

Before using secondary data, the researcher should examine their relevance, reliability, accuracy, completeness, and timeliness.

Secondary data should not be accepted without critical evaluation. The following criteria help determine whether the data are suitable:

• Suitability: The concepts, definitions, units, and population covered should match the current research objectives.
• Adequacy: The data should provide sufficient detail and coverage for the intended analysis.
• Reliability: The credibility of the original source, method of collection, sampling procedure, and possibility of error should be assessed.
• Accuracy: The researcher should examine how carefully the data were measured, recorded, edited, and presented.
• Timeliness: Recent data are generally preferable when studying changing social, economic, or technological conditions.
• Completeness: Missing observations, incomplete categories, and gaps in the time period should be identified.
• Consistency: Definitions and measurement procedures should remain consistent across sources or time periods.
• Objectivity: The purpose and possible bias of the organization that collected the data should be considered.

Using unsuitable secondary data may produce misleading findings even when the analysis is technically correct.

Primary and secondary data differ in their origin, purpose, collection process, and suitability.

The choice depends on the objectives, available resources, required accuracy, and availability of suitable existing information.

Qualitative research is an interpretive approach used to understand meanings, experiences, perceptions, relationships, and social processes.

Its main characteristics are:

• Exploratory: It investigates relatively unknown or complex issues.
• Contextual: Behavior and experiences are studied within their natural social, cultural, or organizational settings.
• Meaning-oriented: It focuses on how participants interpret their experiences.
• Flexible: Research questions, sampling, and data collection may be refined as understanding develops.
• Inductive: Concepts and explanations are commonly developed from the collected data rather than imposed in advance.
• Researcher involvement: The researcher is an important instrument in collecting and interpreting data.
• Holistic: It examines situations as connected wholes rather than isolating variables alone.
• Textual or visual evidence: Data may consist of interview transcripts, field notes, documents, photographs, audio, or video.
• Small purposive samples: Participants are selected because they can provide rich and relevant information.

Qualitative research seeks depth and understanding rather than statistical generalization.

Qualitative data can be collected through several methods that enable detailed exploration of participants' views and behavior.

• In-depth interviews: Open-ended conversations that explore personal experiences, beliefs, and interpretations.
• Focus group discussions: Guided discussions among a small group of participants that reveal shared and conflicting opinions.
• Participant observation: The researcher observes and may participate in the activities of the group being studied.
• Non-participant observation: The researcher observes events without actively taking part in them.
• Case studies: Detailed investigation of a person, group, institution, event, or community.
• Document analysis: Examination of diaries, letters, policy documents, reports, social media content, and other written materials.
• Life histories and narratives: Collection of personal accounts to understand experiences over time.
• Audio-visual methods: Use of photographs, recordings, videos, and other visual materials.

The selected method should match the research question, setting, ethical requirements, and level of access to participants.

Quantitative research is a systematic approach that collects numerical data and uses statistical procedures to describe phenomena, test hypotheses, and examine relationships among variables.

Its major characteristics include:

• Measurement: Concepts are operationalized into observable and measurable indicators.
• Structured design: Instruments, procedures, and analysis plans are usually specified before data collection.
• Deductive reasoning: Existing theories or hypotheses are tested using empirical evidence.
• Numerical data: Findings are expressed through counts, scores, percentages, averages, and other numerical measures.
• Large or representative samples: Probability sampling may be used to improve generalizability.
• Objectivity: The researcher attempts to minimize personal influence on measurement and analysis.
• Statistical analysis: Descriptive and inferential statistics are used to analyze data.
• Replication: Clear procedures make it possible for other researchers to repeat the study.
• Hypothesis testing: Relationships, differences, and effects may be examined systematically.

Quantitative research is especially appropriate when the aim is measurement, comparison, prediction, or generalization.

Qualitative and quantitative research are distinct but complementary approaches.

Neither approach is universally superior. The appropriate choice depends on the research question and the type of knowledge required.

In qualitative research, data are the materials from which the researcher develops understanding. They may include interview transcripts, field notes, observations, documents, photographs, recordings, and personal narratives.

A variable is a characteristic that can take different values across cases. In quantitative research, variables are usually measured numerically. In qualitative research, the idea of a variable is used more flexibly because the focus is often on concepts, meanings, categories, and processes rather than numerical scores.

For example, in a study of student learning:

• Interview statements about motivation are qualitative data.
• Repeated ideas such as fear of failure, peer support, or self-confidence are categories or themes.
• These themes may represent conceptual dimensions related to the broader concept of motivation.

Qualitative researchers generally do not reduce all evidence to fixed variables at the beginning. Instead, they may allow concepts and categories to emerge during analysis. The emphasis is on depth, context, relationships, and interpretation. However, qualitative studies can still compare cases according to characteristics such as age group, role, location, or experience when such distinctions are relevant.

The development of categories and themes is a central part of qualitative data analysis.

• Familiarization: The researcher reads or reviews the data repeatedly to understand its overall content.
• Initial coding: Important words, phrases, actions, or ideas are labeled with short codes.
• Comparison: Similar and contrasting codes are compared across interviews, observations, or documents.
• Category formation: Related codes are grouped into broader categories that represent a common idea.
• Theme development: Categories are connected to form themes that explain important patterns or meanings in the data.
• Review: Themes are checked against the original data to ensure that they are coherent and well supported.
• Definition: Each theme is clearly named and described so that its boundaries are understood.
• Interpretation: The researcher explains how the themes answer the research questions and relate to the study context.

For example, codes such as "lack of time," "work pressure," and "family responsibilities" may form the category competing demands, which may contribute to the wider theme barriers to participation.

Quantitative research uses numerical data organized through variables. A variable is a measurable characteristic that can take different values among individuals, objects, or situations.

Common types include:

• Independent variable: The presumed cause, predictor, or condition that influences another variable.
• Dependent variable: The outcome or effect that is measured.
• Control variable: A factor held constant or statistically controlled to reduce alternative explanations.
• Extraneous variable: An additional factor that may affect the dependent variable but is not the main focus.
• Discrete variable: A variable with countable values, such as number of visits.
• Continuous variable: A variable that can take any value within a range, such as height or income.
• Categorical variable: A variable whose values represent groups or categories.
• Numerical variable: A variable expressed through numbers and measurable quantities.

Variables may also be classified by measurement scale as nominal, ordinal, interval, or ratio. Clear operational definitions are necessary so that variables can be measured consistently.

The four levels of measurement determine the meaning of values and the statistical operations that are appropriate.

• Nominal scale: Values are labels or categories with no meaningful order. Examples include gender category, religion, or type of organization. Frequencies and modes are appropriate.
• Ordinal scale: Values have a meaningful order, but the differences between ranks are not necessarily equal. Examples include satisfaction levels or class position. Medians, percentiles, and rank-based comparisons may be used.
• Interval scale: Values are ordered and have equal intervals, but there is no true zero. Temperature measured in Celsius is an example. Addition and subtraction are meaningful.
• Ratio scale: Values have equal intervals and a meaningful zero, allowing statements about absolute quantities and ratios. Examples include age, weight, income, and duration. All common arithmetic operations are possible.

The scale of measurement affects how data should be summarized and which statistical tests can be applied. Treating an ordinal or nominal variable as if it were a ratio variable may lead to invalid conclusions.

Variables play different roles in explaining a research relationship.

• Independent variable: The factor expected to influence another variable.
• Dependent variable: The outcome that is expected to change because of the independent variable.
• Intervening or mediating variable: A factor that explains the process through which the independent variable affects the dependent variable.
• Control variable: A factor kept constant or included in analysis so that its influence does not confuse the main relationship.

Example: Suppose a researcher studies whether training improves employee productivity.

• Training hours are the independent variable.
• Employee productivity is the dependent variable.
• Employee skill development may be an intervening variable, because training may improve skills, which then improve productivity.
• Previous work experience, department, or working hours may be treated as control variables.

Correctly identifying variable roles helps researchers formulate hypotheses and select appropriate methods of analysis.

Operationalization is the process of converting an abstract concept into specific, observable, and measurable indicators.

For example, the concept of academic achievement may be operationalized through examination scores, grade-point average, course completion, or standardized test results. The concept of job satisfaction may be measured through responses to several statements using a rating scale.

The process generally involves:

• Defining the concept clearly.
• Identifying its dimensions.
• Selecting observable indicators for each dimension.
• Choosing an appropriate measurement scale.
• Developing questions, items, or procedures for measurement.
• Specifying how scores will be combined and interpreted.

Operationalization is important because it:

• Makes abstract concepts measurable.
• Improves consistency among researchers.
• Supports reliable data collection.
• Allows comparison across individuals and groups.
• Helps test hypotheses empirically.
• Makes replication of the study possible.

Poor operationalization can create measurement error and weaken the validity of the research findings.

Qualitative data collection and management require systematic planning even when the research design is flexible.

• Develop the research focus: Identify the issue, participants, setting, and broad research questions.
• Select participants: Use purposive, theoretical, snowball, or other suitable sampling strategies.
• Obtain informed consent: Explain the purpose, procedures, risks, confidentiality, and voluntary nature of participation.
• Collect rich data: Conduct interviews, observations, discussions, or document reviews using appropriate protocols.
• Record the context: Maintain field notes about the setting, interactions, non-verbal behavior, and researcher reflections.
• Transcribe and organize: Convert recordings into transcripts and assign identifiers to participants and cases.
• Protect confidentiality: Remove identifying information and store files securely.
• Code and analyze: Mark meaningful sections, develop categories, compare cases, and identify themes.
• Maintain an audit trail: Record decisions about sampling, coding, interpretation, and changes in the research process.

Good data management improves transparency, credibility, and the dependability of qualitative findings.

Validity refers to the extent to which a measurement instrument measures what it is intended to measure. Reliability refers to the consistency or stability of the measurement.

Types of validity include:

• Face validity: Whether the instrument appears appropriate on the surface.
• Content validity: Whether the instrument covers all important aspects of the concept.
• Construct validity: Whether the measurement represents the theoretical concept it is intended to measure.
• Criterion-related validity: Whether scores are related to an appropriate external criterion.

Types of reliability include:

• Test-retest reliability: Consistency of scores over time.
• Internal consistency: Agreement among items measuring the same concept.
• Inter-rater reliability: Agreement between different observers or coders.
• Parallel-forms reliability: Consistency between equivalent versions of an instrument.

An instrument may be reliable without being valid if it produces consistent but inaccurate measurements. Researchers should pilot instruments, use clear items, standardize procedures, and select established measures where possible.

Trustworthiness is used in qualitative research to assess the quality and credibility of interpretations. Its major criteria are:

• Credibility: The confidence that the findings accurately represent participants' experiences. It may be strengthened through prolonged engagement, member checking, triangulation, and peer debriefing.
• Transferability: The extent to which findings may be relevant in other contexts. Researchers support transferability by providing rich descriptions of the participants, setting, and research process.
• Dependability: The consistency and logical stability of the research process. An audit trail and clear documentation of methodological decisions are useful.
• Confirmability: The extent to which findings are grounded in the data rather than researcher preferences. Reflexive notes, audit trails, and evidence such as quotations support confirmability.

Additional practices include seeking negative cases, comparing multiple data sources, discussing interpretations with peers, and clearly acknowledging researcher positionality. These procedures do not make qualitative research identical to quantitative research; they provide appropriate standards for evaluating interpretive inquiry.

Writing up qualitative research involves presenting the study's context, process, evidence, and interpretation in a coherent manner.

• Introduction: Present the research problem, context, purpose, and research questions.
• Literature review: Explain relevant concepts, previous studies, and the gap addressed by the research.
• Methodology: Describe the research approach, setting, participants, sampling, data collection, ethical procedures, and method of analysis.
• Researcher reflexivity: Explain the researcher's position and possible influence on the study.
• Findings: Present major categories and themes supported by carefully selected participant quotations, observations, or documents.
• Interpretation: Explain the meaning of the findings and connect them to the research questions and existing literature.
• Negative or contrasting evidence: Report important differences and cases that do not fit the dominant pattern.
• Quality and limitations: Discuss credibility procedures, study limitations, and the boundaries of interpretation.
• Conclusion: Summarize the main insights, implications, and possible areas for further research.

The report should preserve participants' meanings while maintaining analytical clarity, ethical confidentiality, and a clear distinction between evidence and interpretation.

Thematic analysis is a method of identifying, organizing, analyzing, and reporting recurring patterns of meaning, known as themes, within qualitative data.

A common process includes:

• Data familiarization: Read transcripts, field notes, or documents repeatedly.
• Generating initial codes: Label relevant passages with concise descriptions.
• Searching for themes: Group related codes into possible themes.
• Reviewing themes: Check whether the themes are internally coherent and adequately supported by the data.
• Defining and naming themes: Describe the central idea and boundaries of each theme.
• Writing the report: Present the themes with evidence and interpret their significance.

For instance, interview data on remote work may produce codes such as isolation, flexible schedules, and blurred work boundaries. These may be grouped into themes such as changing work-life boundaries and social consequences of remote work.

The researcher should remain close to the data, document analytical decisions, consider alternative interpretations, and avoid selecting quotations that misrepresent participants' views.