TLDR;
This video provides a detailed explanation of research methodology, focusing on hypothesis formulation, data collection, sampling techniques, and data analysis. It differentiates between qualitative and quantitative data analysis methods, parametric and non-parametric statistics, and various experimental designs. The lecture also covers the essential steps in conducting research, tailored to different research types such as philosophical, historical, descriptive, and experimental research.
- Hypothesis formulation requires consideration of tense and appropriate terminology.
- Data collection methods vary based on the type of research (philosophical, historical, survey, experimental).
- Sampling techniques are divided into probability and non-probability sampling, each with specific methods suitable for different population characteristics.
- Data analysis differs for qualitative and quantitative data, with specific statistical techniques applicable based on the scale of measurement and data distribution.
Introduction [0:00]
The session begins with an invitation for participants to ask questions related to the material covered before the lunch break. The presenter indicates readiness to address any queries before moving forward with new content.
Hypothesis Formulation and Testing [0:36]
The discussion transitions to the formulation of hypotheses, emphasising the importance of tense when writing a hypothesis. Hypotheses should be written in the present tense and include words suggesting doubt or significance. The presenter also touches on hypothesis testing, introducing the concepts of degrees of freedom and levels of significance as essential for testing hypotheses. Two types of tests, one-tail and two-tail tests, are explained in relation to null and directional hypotheses. The lecture also mentions Type I and Type II errors, highlighting their importance in research.
Data Collection [4:58]
The lecture moves to data collection, differentiating between quantitative and qualitative data and when to formulate a hypothesis. Data collection methods vary depending on the type of research being conducted. Philosophical research requires primary sources, historical research uses primary and secondary sources, while survey, experimental, action, and case study research involve collecting data from the field or through experiments in laboratories. The presenter stresses the importance of understanding sampling, instruments, and research methodologies when collecting data from the field or lab.
Sampling Techniques [8:09]
The presenter discusses sampling, covering concepts like universe, population, sample, frame, and sampling techniques. Sampling techniques are categorised into probability and non-probability sampling. Probability sampling includes various techniques depending on whether the population is widely spread or homogeneous. If the population is not widely spread and homogeneous, random sampling can be used. If the population is not widely spread but heterogeneous, stratified random sampling is appropriate. For populations that are widely spread, cluster sampling or multi-stage sampling techniques are recommended. Non-probability sampling techniques include purposive, judgment, convenience, quota, and snowball sampling.
Sample Size and Tools [19:14]
The lecture addresses determining sample size and differentiating between small and large samples. It also covers the difference between statistics and parameters. The quality of data depends on the quality of the tools used. Researchers need to understand the characteristics of their instruments, distinguishing between measurement and evaluation, and standardised versus non-standardised tools. Standardisation requires quantitative data, and non-standardised tools may yield data that is not widely trusted, affecting the credibility of research findings.
Research Methods [26:10]
The presenter outlines various research methods, including philosophical, historical, survey, and experimental methods. Experimental methods require understanding components of experiments, experimental designs, and their principles. Experimental designs are classified into pre-experimental, quasi-experimental, and true experimental designs. Pre-experimental designs are considered the weakest. True experimental designs are suitable for science-related research, while quasi-experimental designs are more appropriate for social science research. The lecture also touches on internal and external validity and factors affecting them.
Data Analysis [35:37]
The discussion transitions to data analysis, which is essential for drawing conclusions. The approach to analysis depends on whether the data is qualitative or quantitative. Qualitative data can be analysed using content analysis, meta-analysis, or SWOT analysis. Quantitative data analysis requires understanding scales of measurement to select appropriate statistical techniques, which fall under parametric and non-parametric statistics.
Parametric vs Non-Parametric Statistics [41:58]
The presenter explains the differences between parametric and non-parametric statistics. Parametric statistics are used when data has a normal distribution and is measured on an interval or ratio scale. Non-parametric statistics are used when data does not have a normal distribution or is measured on a nominal or ordinal scale. Parametric statistics are more powerful. Examples of parametric statistical techniques include t-tests, ANOVA, correlation, and regression analysis. Non-parametric techniques include chi-square tests, median tests, sign tests, and rank correlation methods. Non-parametric statistics are also known as distribution-free statistics.
Research Steps Summary [59:26]
The lecture summarises the steps in research, which include identifying a problem, collecting data, analysing data, and drawing conclusions. For qualitative research, the steps include problem identification, data collection (using primary and secondary sources), data analysis (using methods like content analysis), and conclusion. For quantitative research, the steps include problem identification, hypothesis formulation, data collection (including sampling and instrumentation), data analysis, and conclusion.
Hypothesis Clarification [1:07:17]
The presenter clarifies that researchers should not provide both research and statistical hypotheses, as it is technically incorrect. However, if a guide or supervisor insists, it is advisable to comply.
Problem Identification and Variables [1:08:53]
The session moves to the first step in research: problem identification, focusing on variables. A variable is defined as a changeable characteristic. Examples of variables include weight, height, reading comprehension, vocabulary, confidence, teaching medium, colour, blood pressure, interest, personality, social problems, marital status, age, cost, subject, and more.
Constants vs Variables [1:25:05]
The presenter contrasts variables with constants, which are unchangeable characteristics. Examples of constants include the number of days in a week, the value of pi, and the sum of angles in a triangle. A variable can be held constant for a specific purpose or time, but a constant cannot be made variable. The term 'held constant' is preferred over 'made constant'.