Research Methodology Lecture Notes PDF

# Lecture 1 ## Research: Methodology to Solve Problems, New Knowledge, Applicable **Reasons:** - Understand - Explain, facilitate knowledge transfer - Examine aspects - Make change & promote good practice - Find answers to theoretical questions - Policy suggestions & stimulate social innovation...

# Lecture 1 ## Research: Methodology to Solve Problems, New Knowledge, Applicable **Reasons:** - Understand - Explain, facilitate knowledge transfer - Examine aspects - Make change & promote good practice - Find answers to theoretical questions - Policy suggestions & stimulate social innovation ## Research Question: A research question is a clear, focused question that guides your study. It defines the scope and direction of your research, helping to determine what you will investigate. ## Types of Research: | Type of Research | Viewpoint | Objectives | Enquiry Mode | |---|---|---|---| | Pure Research | Application | Descriptive | Mixed Methods | | Applied Research | Objectives | Exploratory | Quantitative | | | | Correlational | Qualitative | | | | Explanatory | ## Pure Research: 1. Developing, testing theories, hypotheses (contains abstract) 2. Development, examination, verification & refinement of research methods, procedures, techniques & tools that form body of research methodology 3. May/not practical application present/future (sampling method, instrument, validity) ## Applied Research: Techniques, procedures, methods forms body with information (evidence-based decision making, policy formulation, enhancement understanding of phenomenon) ## Descriptive Study: 1. Describe what is prevalent. 2. Situation, problem, phenomenon, service, program ## Correlational Study: Discover relationships between 2/+ aspects of situation ## Explanatory Research: Clarify why & how. ## Exploratory Research: Explore possibilities of undertaking research study ## Quantitative Approach: (Number based, measurement, large sample size) 1. Everything forms research process is predetermined (objectives, design, sample, questions). 2. Determine problem. 3. Structured procedures. 4. Emphasize objectivity of process. 5. Importance to validity & reliability. ## Qualitative Approach: (Quality, descriptive, fewer case) 1. Flexible. 2. Explore variation in issue. 3. Importance to validity & reliability. ## Mixed Method Approach: Qualitative & quantitative. ## Convergent Design: * **Quantitative Data Collection & Analysis** * **Qualitative Data Collection & Analysis** * **Compare or relate** * **Interpretation** ## Explanatory Sequential Design: **Quantitative results = Qualitative data & results & Qualitative exploration = Quantitative test** * **Quantitative Data Collection & Analysis** * **Follow up with** * **Qualitative Data Collection & Analysis** * **Interpretation** * **Qualitative Data Collection & Analysis** * **Builds to** * **Quantitative Data Collection & Analysis** * **Interpretation** # Lecture 2 ## Function of Literature Review: 1. Main findings - organized around main themes. 2. Themes are in logical manner. 3. Findings compared (similarities & differences). ## How to Read a Literature Review: 1. Understand main themes & how findings structured. 2. Look for logical flow of arguments & evidence provided. ## Writing a Literature Review: 1. Thematic. 2. Logical order. 3. Arguments referenced evidence from existing literature. 4. Academic Referencing Style. ## Literature Summary & Review Difference: * **Summary**: No logic, no theme. * **Review**: Thematic, based main theme, theme based analysis, logical order. ## What are the Requirements of a Research Proposal - APA ## How to Select a Subject: Experimental group & control group are as similar as possible. ## Covariation: * **Positive Correlation** * **Negative Correlation** * **No Correlation** ## Spurious Relationship: * **Cause** (independent variable) * **Effect/Outcome** (dependent variable) * **Other Factors** (confounding variable) ## Experiments: * Experimental, non-experimental research designs * Controlled experiment, research method associated with natural sciences * Experiments often used in nonscientific human inquiry * Experimental model especially appropriate for hypothesis testing, focusing on determining causation * Laboratories, internet, regular course of social events ## Classical Experiment: 1. **Independent & Dependent Variables**: Examines effect of independent variable on dependent variable. 2. **Independent Variable Takes for of Experimental Stimulus (Present/Absent)**: Compares what happens when stimulus is present/when it's not. 3. **To be used in experiment both must be operationally defined.** 4. **Pretesting & Posttesting**: Subjects are measured in terms of a dependent variable. The subjects are exposed to a stimulus representing an independent variable. Subjects are then remeasured in terms of the dependent variable. ## Experimental & Control Group: * **Experimental**: Group of subjects whom an experimental stimulus is administered. * **Control Group**: Group of subjects whom no experimental stimulus is administered, who resemble experimental group in other respects. * **Comparison of both group at the end points to the effect of the experimental stimulus.** ## Treatment Effect: **Net Treatment Effect = (TreatM2 - TreatM1) - (ControlM2 - ControlM1)** ## Placebo Effect: Participants reporting real improvement after taking a fake treatment. ### Considerations Without a Control Group: Challenging to determine cause & effect relationship of experimental treatment, leading to confounding variables affecting results, making it difficult to draw valid conclusions. ## What We Want to Achieve in Experiment: Test hypothesis & establish causal relationships between variables, understanding underlying mechanisms of phenomena. ## Reliability: Some data collected in repeated observations. The extent of getting same results if we repeat the measurement. ## Validity: Measure accurately reflects the concept intended to measure. ### Internal Validity: Research procedure demonstrates real cause & effect relationship. ### Internal Invalidity: Possibility conclusion drawn not accurate (sources: history & testing) ### External Invalidity: Possibility conclusions drawn from experimental results may not be generalizable to "real" world (populations, times, settings). ## Potential Problems in Validity & Reliability: Credibility of findings. ## Randomization: Technique used to assign subjects to experimental & control group randomly (eliminate bias) ## Advantages & Disadvantages of Close Ended Questions (Eg: MCQ) * **Advantages**: Easy to answer & analyse. * **Disadvantages**: Information lacks depth & variety, greater possibility of bias, may not reflect respondents' opinion. ## Advantages and Disadvantages of Open Ended Questions: * **Advantages**: In-depth & variety information, no investigator bias. * **Disadvantages**: More challenging to analyze (answers need to be classified), loss information if respondents can't express themselves, possible interviewer bias. # Types of Survey Method: ## Self-Administered Questionnaires: * Respondents complete themselves. * Mail survey. * Home delivery of questionnaires. ## Interview Surveys: * Send interviewers to ask questions orally & record answers. * Interviewer must be neutral. * Interviewer must be familiar with the question, follow the steps. * Interviewers can use probes to elaborate on incomplete response. ## Telephone Surveys: * Cheaper than face to face. * More honest. * Random digit dialing is useful for eliminating bias. * Computer assisted telephone interviewing: Questionnaire stored in computer, permitting interviewer to read and enter the answers on keyboard. ## Online Surveys: * Surveys over internet. * Respondents may not be representative. ## Comparison of Different Survey Methods: 1. **Self-administered questionnaires over interview survey**: Economy, speed, lack interviewer bias, possibility of privacy to encourage responses on sensitive issues. 2. **Interview survey over self-administered questionnaire**: Fewer incomplete questionnaires, fewer misunderstood question, flexibility in sampling & special observations. 3. **Telephone survey over face to face interviews**: Saving cost & time, safer than in person interview. ## What is Sampling: The process of selecting individuals from a population to participate in a study. ## Two Major Types of Sampling Design: ### **Probability Sampling**: * Simple random * Stratified * Systematic * Cluster ### **Non-Probability Sampling**: * Quota * Purposive * Snowball * Convenience ## 1. Simple Random Sampling: Each member has an equal probability of being chosen. This method involves preparing a list of all population members and selecting randomly from it. ## 2. Stratified Sampling: * The population is divided into specific sub-groups (strata) based on characteristics and samples are drawn from each stratum. * Samples taken from everyone in group, only some members in each group selected. * Presentation better. ## 3. Cluster Sampling: * This method is used when the population is divided into clusters, and entire clusters are randomly selected. * Natural groupings, geographical proximity. * Members in each cluster need to be different, each cluster = mini presentation of entire population. * All members of selected clusters included. * Sample lack representations. ## 4. Systematic Sampling: Involves numbering the population and selecting every nth individual after a random start point. ## When to Use: * **Simple random sampling**: Whole population available. * **Stratified sampling**: There are specific sub groups to investigate. * **Cluster sampling**: When population groups separated & access is difficult. * **Systematic sampling**: Cheaper of simple random sampling, use systematic when on tight budget/short timeline. # Non-Probability Sampling: Don't require equal opportunity in sample selection, personal bias might be involved. ## 1. Quota Sampling: Main consideration: ease access to sample population & visible to characteristic. ## 2. Convenience/Accidental Sampling: Samples are taken from a group that is easily accessible. ## 3. Purposive/Judgemental Sampling: Researcher selects individuals based on their judgment about who would be most representative. ## 4. Snowball Sampling: Selecting sample using network (useful when population unknown/rare/unwilling to be identified). ## Differences in Sampling: Quantitative & Qualitative: ### **Quantitative:** * Unbiased * Representative * Make inferences * Random samples * Large sample size ### **Qualitative:** * Ease of access * Purpose driven * Information rich respondents * A few cases until saturation is reached. ## Probability Sampling: Involves random selection ensuring each individual has a known chance of being chosen. ## Non-Probability Sampling: Doesn't guarantee, can introduce bias. ## Participant Observation: Participate in activities of groups being observed with/without knowing they're being observed. ## Non-Participant Observation: Don't get involved, passive, only watching & listening then draw conclusion. ## Methods Triangulation: Involves integrating multiple methods in analyzing information. ### Cross Method Triangulation: Different methods used simultaneously. ### Within Methods Triangulation: Multiple analysts examine same data to reduce bias. ## Data Triangulation: Expands data base by collecting additional data. * **Time Triangulation**: Collecting at various points in time to check for consistent & identify anomalies. * **Space Triangulation**: Gather data from multiple locations to compare. * **Person Triangulation**: Obtaining information from various respondents.