Research Methods and Data Analysis

Module 1: Qualitative Research Designs

• Qualitative research can be challenging for new researchers.
• There are various qualitative research types including ethnography, phenomenology, grounded theory, case study, narrative, and historical research.
• Selecting the right approach is key to answering the research question.
• Qualitative research design is an umbrella concept covering a range of methods with varying focus and assumptions.
• Qualitative research is helpful for answering "How?" and "What?" questions.
• Qualitative research design involves alternate terms like 'research methodology', 'research approach' and 'research type'.
• Research design is the transformation of a research idea into a practical research project.

Different Qualitative Research Designs:

• Ethnography provides a direct group, culture, or community description; data is collected via fieldwork, interviews, participant observation, and documents.
• Researchers observe cultural rituals while trying to understand the meanings from informants.
• Comparisons are made with researchers’ ethnic views to explore differences.
• Field notes are recorded throughout fieldwork.

Grounded Theory (GT)

• Grounded Theory is an approach that generates and modifies a theory.
• The end product is also referred to as a GT–a development of theory directly based on data collected by the researcher.
• Data collected from non-standardized interviews, participant observation, and other data sources.
• Researchers code and categorize transcripts from interviews or field notes.
• Dialogue is used with the literature when discussing categories; constant comparison and theoretical sampling are also implemented.
• Memos—theoretical notes—help develop theoretical ideas.
• The theory generated has exploratory power and is grounded in data.

Narrative Inquiry

• Narrative Inquiry includes stories reflecting peoples’ experiences and meanings.
• This research is useful for accessing and analyzing feelings, thoughts, and experiences.
• Narratives are inherently tales of experience or imagination.
• Researchers conduct in-depth interviews, review documents, and identify themes to see how an individual's story reflects larger life influences.

Phenomenology

• Phenomenology focuses on the subjective experience of the individual.
• Its purpose is to describe lived experiences.
• Data collected typically through in-depth conversations.
• Researchers assist participants in describing lived experiences avoiding leading the discussion.

Case Study

• The case study design provides an in-depth look at one test subject, which can be a person, family, business, organization, or town.
• Interpretative research is used, since researchers are interested in insight and discovery.
• Multiple data collection methods are employed like interview, observation, and document review
• A final report is generated at the end of the study.

Module 2: Sampling Procedure

• Sampling means choosing a subgroup from a population to participate in a study, so the sample represents the larger population.
• Probability and non-probability sampling make up the two major sampling procedures.

Probability Sampling

• Everyone has an equal chance of being selected
• Every unit has a greater-than-zero chance of being selected.
• Basic types of sampling procedures are:
  • Simple random
  • Systematic
  • Stratified
  • Cluster

Simple Random Sampling

• Simple random sampling serves the basis for complex sampling methodologies
• Prepare an exhaustive list/sampling frame of members within the interest population.
• The sample is drawn from this list giving everyone a fair chance to be drawn during each selection round.

Systematic Sampling

• Involves selecting every nth member, randomly selecting the first through nth element as the starting point.
• Example:
  • To sample 20 respondents from 100, every 5th person is selected.
• Systematic samples tend to be easier to draw/execute.
• Researchers do not have to go back and forth through the sampling frame.
• The selection might spread members selected more evenly across the population resulting in a more representative sample.

Stratified Sampling

• The most effective sampling method when wanting a representative sample of a population.
• Categorizes population members into mutually exclusive/collectively exhaustive groups.
• An independent simple random sample is drawn from each group.
• Stratified techniques can provide precise estimations if the population is more heterogenous than categorized groups.
• It can determine desired levels of sampling precision and provide administrative efficiency.
• Can give the most representative sample.

Cluster Sampling

• Involves a cluster (group of population elements) constituting the sampling unit instead of a single population element.
• A cluster unit for sampling is mainly geographically driven.
• Cost efficiency (economy and feasibility) is the main reason.
• The sampling frame is readily available at a cluster level and takes shorter time for listing/implementation.
• Suitable for surveys of institutions or households in a geographical area.
• Disadvantages:
  • May show a poor reflection of community diversity
  • Elements sharing similar characteristics
  • Provides less information than an SRS of the same size
  • Higher standard error of estimates compared to sampling designs with the same size.

Non-Probability Sampling

• Used in situations where the population is not well defined or interest in drawing inferences from the sample to the population is low.
• Reason for use is less expensive, quicker implementation.
• Includes:
  • Purposive
  • Convenience
  • Voluntary
  • Snowball sampling

Purposive/Judgmental Sampling

• The researcher chooses the sample based directly on who would be appropriate for the study to adequately answer research objectives.
• Selection involves applying expert knowledge of the target population in a non-random manner.
• Disadvantage:
  • Subjectivity, a different researcher might not be able to replicate the sample as accurately (identifying important characteristics and picking typical elements for the sample)
• Most appropriate for small samples from limited geographic areas.

Convenience Sampling

• Also known as opportunity, accidental or haphazard sampling.
• Involves a sample drawn from a part of the population close to hand or readily available/convenient.
• This sample type is useful for pilot testing.
• Differs from purposive sampling since expert judgement is not used to select a representative sample.
• Primary selection criterion is ease of obtaining a sample based on the cost of locating elements, sample geographic distribution, and obtaining data from members of population.

Voluntary Response Sampling

• A convenience sample based on easy access; participants volunteer themselves instead of the direct researcher selecting.
• These are somewhat biased, likely due to inherent volunteer propensity.

Snowball Sampling

• Used when the population is hard to access, recruiting participants via other participants.
• The number increases as contact is made.

Module 3: Data Collection, Data Gathering Instrument, and Analysis Procedure

• After researching and clarifying the framework/perspective, gather supplementary evidence.
• Primary sources provide raw data for analysis.
• Planning methods becomes more focused and intentional for appropriate questions.
• Qualitative research goal is usually to explore, describe, or interpret a situation, phenomenon, social issue, or reality.
• It is also used as more open-ended methods, such as:
  • Observations
  • Interviews
• Emphasizes that reality is socially constructed and could be interpreted, so there is an intimate connection between researcher/situation.
• Qualitative data collection approaches usually involve direct interaction with individuals, one-on-one or in group settings.
• This data collection method takes up time; data collected from smaller samples leading to a more expensive qualitative study although information has deeper insight.

Data Collection Methods

• Individual interviews
• Focus group
• Observations
• Action Research

Observation

• Systematic, intentional process using the researcher to observe phenomena, interactions, issues, or realities
• 2 types of observations:
  • Participant observation occurs when the researcher participates in the group activities/acts like the community to assimilate
  • In non-participant observation, the researcher is a passive community member.
• Both cases require consent and enough time to assimilate so people can be observed.

Interview

• Method of collecting information about people, their thoughts, behaviors, opinions, concerns, or aspirations.
• Person-to-person purposeful interaction for relevant, insightful information.
• Interviews include:
  • One-time
  • Multiple with same participants
  • Multiple with various participants
  • Group interviews
• Can be done through:
  • Face-to-face interactions
  • Telephone
  • Focused group discussions
  • Online conferencing platforms

Interview Types

• Unstructured, using open-ended questions
• Structured using close-ended questions
• Semi-structured using both the above

Questionnaire

• More formal way to gather facts, data, opinions through well-constructed instruments with very clear purpose and directions, in a printed format.
• Written list of questions filled out by the respondent and then given back to the researcher.
• This involves respondents replying to open/closed-ended questions.

Documents

• Poring over and recording notes on public (newspapers, meeting minutes, official reports) or private documents (journals, diaries, letters, emails).

Audio and Visual Materials

• Studying photographs, art objects, videotapes, or forms of sound for relevant themes, similarities, and differences.

Data Analysis

• Coding:
  • The analyst reads data and marks segments; each segment is labeled with a code (word/phrase) suggesting how data segments inform research objectives.
  • Preparation for reporting is via mix summarizing, discussing similarities and differences in related codes, and context comparing the relationship.
• Recursive:
  • Data sets are summarized for a more compact result.
• Content or discourse analysis:
  • Counting words, phrases, or coincidences of tokens; other similar techniques are analyses of conversation phrases
• Computer Assisted Qualitative Analysis Software:
  • enhances efficient data storage, retrieval, and applying data codes. -MAXQDA, QDA MINER, ATLAS ti, Dedoose (mixed methods) and NVivo are common qualitative data analysis softwares.

General Steps in Data Analysis

• Organize and prepare data like interviews, materials, notes and sources of information.
• Read and obtain a general sense of the information and idea tones.
• Begin detailed analysis by organizing text data/pictures to categories and label with terms.
• Use categories & themes for rendering detailed information/descriptions for analysis.
• Theme appears as findings; stated under the separate section of the study.
• Decide how the description will be presented.
• Use narrative passages/detailed discussions of themes, visuals, figures, or tables.
• Make a personal interpretation of meanings derived from a finding comparison/questions for future study which were not foreseen earlier in the study.

Module 4: Research Methodology

• Practical research 1 uses practical skills for idea development
• Multiple-step discipline process is interlinked.
• Serves as an umbrella covering inquiry forms to understand meanings with as little setting disruption.
• Research is a systematic process of logically collecting/analyzing the data for a given purpose
• Research helps acquire new knowledge with planning, investigating, and intervening to interpret the new information that leads to innovation and improvement.
• The methodology is the philosophical framework of the research foundation
• It is designed in detail highlighting use throughout a study, the choice justification describing advantage/disadvantage, and its practical applicability.

Types of Research Designs

• Qualitative Method, Quantitative Method

Qualitative Method

• Case Study sheds light on studying a single case.
• Ethnography is a close observation of a sociocultural phenomenon.
• Researchers immerse themselves in studied cultures; these are lengthy studies.
• Phenomenology attempts to provide understanding people's understanding from the experiences.
• Grounded Theory derives a theory inductively from data acquired, with the goal of developing a theory about the process.

Qualitative Research Methods

• Qualitative research methods collect data at the location the participants are experiencing issues/problems.
• Real-time with rarely having the participants to meet data collection needs.
• Qualitatively gather data using interviews, observations, and documents rather than data sources.
• Work towards solving complex issues, breaking them down into meaningful, readable easy-to-understand inferences.

Quantitative Method

• Experimental Research design
• True Experimental
• Researchers manipulate and identify cause and effect with randomization
• Quasi-Experimental
• Strategy, intervention, or treatment is used.
• Absence of subject random assignment under different conditions.
• Non-equivalent control group design
• Time-series designs
• Pre-Experimental
• One-shot case study
• Implement to a single group
• Apply pre-test, post-test design
• Measure the group before and after intervention
• It has the least internal validity.
• Non. Experimental (Descriptive) Research design
• Survey Research
• Used whenever a researcher provides a quantitative description attitudes from the samples population.
• Correlational Design
• Calculating the coefficients/variable relationship such as:
  • wealthy children
  • well-educated parents
  • high-paid salary as adult
• Comparative Design
• Involves study subject comparison using one or more variables at a single point.
• Evaluative Design
• Judges or assesses something in some way.
• Purpose relates to providing program improvement or effectiveness.