2026-04-02
Every step includes: Literature review
Ends the step with: Scientific report writing (presentation/publication)
Problem must be consistent with the existing knowledge.
Problem must have newness (innovativeness/novelty).
Title: 10 - 15 words that reflect the research problem and objectives. Sometimes, it may include methodological aspects.
Example: “Effect of row planing on growth and yield of wheat in Bangladesh”, “Effect of training on income generation: A socio-psychological analysis using Kirkpatrick’s model”.
Titles should avoid dead words (investigate, study, effect, etc.), colon (as in the avove example) and abbreviations.
Objectives: Specific statements that describe what the research aims to achieve.
Example: To estimate the effect of training on the adoption of row planing among wheat farmers in Barishal region by the end of 2025.
A. Pre-experimental design: no randomization. e.g. one group pretest-posttest design, one group posttest only design, static group comparison design (posttest only with control group)
B. True experimental design: randomization is possible and there is a control group. e.g. pretest-posttest control group design, posttest only control group design
C. Quasi-experimental design: randomization is not possible but there is a control group. e.g. non-equivalent control group design (has pretest), time series design
The level of measurement determines the appropriate statistical analysis and interpretation of data.
Example: “I am satisfied with the training program on row planing for wheat farmers in Barishal region.”
Response:
Probability sampling: Each member of the population has a known and non-zero chance of being selected. e.g. simple random sampling, systematic sampling, stratified sampling, cluster sampling.
Non-probability sampling: The probability of selection is unknown or zero for some members of the population. e.g. convenience sampling, purposive sampling, snowball sampling, quota sampling.
In cluster sampling a number of clusters are randomly selected for data collection.
In quota sampling quotas are set for different subgroups and data is collected until the quotas are filled.
Remember: Field editing and central editing are part of data processing.
Data processing: The steps taken to prepare raw data for analysis, including data cleaning, coding, and organization.
Data analysis: The application of statistical or qualitative techniques to interpret and draw conclusions from processed data. e.g. descriptive statistics, inferential statistics or hypothesis testing, thematic analysis, content analysis, etc.
Software for data processing and analysis: MS Excel, OnlyOffice, R, Python, SPSS, Stata, MATLAB, Minitab, NVivo, etc.
| Decision | H0 is True | H0 is False |
|---|---|---|
| Reject | Type I Error | Correct Decision |
| Accept | Correct Decision | Type II Error |
Validity: The extent to which a research instrument measures what it is intended to measure. e.g. content validity, construct validity, criterion validity, etc.
Reliability: The consistency and stability of a research instrument over time and across different conditions. e.g. test-retest reliability, inter-rater reliability, split-half or internal consistency reliability, etc.
To be contiued in the next session.
Link: https://ruenresearch.com/blogs/scientific-writing.html#/title-slide
Scan and go through the presentation and resources for the next session.