Summary of Research Methods and Development Methodologies
Research Methods & Development Methodologies: A Guide
Introduction
Research methods provide the tools and frameworks researchers use to ask questions, collect evidence, and draw reliable conclusions. This material focuses on core concepts in research design, sampling, ethics, R&D and innovation, and data presentation—tailored for university students in science and technology fields.
Definition: Research methods are the systematic procedures and techniques used to design studies, gather evidence, analyze results, and report findings.
SECTION 1: Research Purpose, Design, and Classification
Why classify research?
Classifying research by purpose (e.g., exploratory, descriptive, explanatory), by method (qualitative, quantitative, mixed), or by design (experimental, quasi-experimental, survey, case study, action research) helps ensure valid, reliable, and appropriate choices at every stage of a study. Classification guides sampling decisions, data-collection tools, and analysis techniques.
Definition: A research design is the overall plan that links theoretical questions to the methods used to collect and interpret data.
Key components of a good research design
- Clear research question and objectives
- Choice of method aligned with question (e.g., experiments for causality)
- Sampling strategy suited to inference goals
- Data collection instruments with demonstrated quality
- Predefined procedures for analysis and interpretation
Table: Purpose vs Typical Design
| Purpose | Typical designs | Typical outcomes |
|---|---|---|
| Exploratory | Case studies, focus groups, pilot studies | Hypotheses, refined questions |
| Explanatory (causal) | Experiments, quasi-experiments | Cause-effect claims |
| Descriptive | Surveys, observational studies | Profiles, distributions |
| Evaluative | Program evaluation, action research | Practical recommendations |
Practical example
- A technology education researcher wants to know if a new lab manual improves student troubleshooting skills. A randomized controlled trial assigns some classes the new manual (treatment) and others the current manual (control) to measure learning gains and attribute causality.
SECTION 2: Sampling Techniques — Key Comparisons and Guidance
Stratified sampling vs Quota sampling
- Stratified sampling: population is divided into strata based on characteristics; random samples are drawn from each stratum proportional or equal to size. Produces probability-based inferences when sampling frames exist.
- Quota sampling: researcher sets quotas for subgroups and selects participants non-randomly until quotas are met. Easier in the field but introduces selection bias.
Table: Stratified vs Quota
| Feature | Stratified sampling | Quota sampling |
|---|---|---|
| Selection mechanism | Random within strata | Non-random convenience within quotas |
| Sampling frame required | Usually yes | Not necessary |
| Bias risk | Low when executed properly | Higher selection bias |
| Use case | Surveys needing representativeness | Rapid fieldwork, limited resources |
Example: To survey engineering students by year, stratified sampling randomly selects students from each year list. Quota sampling asks interviewers to find a set number of students per year in campus common areas.
Stratified sampling vs Cluster sampling
- Stratified: strata are homogeneous groups (e.g., years of study); sampling within strata improves precision.
- Cluster: population is divided into clusters (e.g., tutorial groups) and whole clusters or random elements from clusters are sampled; efficient when a sampling frame of individuals is costly.
Table: Stratified vs Cluster
| Feature | Stratified | Cluster |
|---|---|---|
| Group purpose | Reduce variability within strata | Reduce cost by sampling groups |
| Homogeneity | Strata are internally similar for key variable | Clusters ideally heterogeneous internal |
Sign up for the full summaryAlready have an account? Sign in
Research Methods Overview
Klíčové pojmy: Research design links questions to methods and drives validity, Stratified sampling uses random selection within homogenous strata, Quota sampling meets subgroups non-randomly and can introduce bias, Cluster sampling samples groups to reduce cost but may lower precision, Qualitative methods explore processes and context; quantitative tests hypotheses, Use Cooper’s Stage-Gate steps to move lab innovations to market, Observe ethical principles: respect, beneficence, justice, confidentiality, Visualize data first to detect errors and patterns, Choose sampling based on resources, required representativeness, and available frames, Pilot qualitative work to improve instruments for quantitative studies, R&D creates knowledge; innovation converts it into market value