Scribd
Upload
78 views

Research Design - Formulating The Research Problem

This document discusses research design and methodology. It explains that research design is a structured plan for how to conduct research, while methodology refers to the specific techniques and procedures used. There are three main purposes of research: exploration to develop an initial understanding, description to precisely measure characteristics, and explanation to determine why something occurs. The document outlines different research methods that align with each purpose, such as interviews for exploration and experiments for explanation. It also discusses key concepts like correlation, causation, units of analysis, and threats to validity that must be considered when formulating the research question and choosing a research design.

Uploaded by

Abby Accad
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
78 views

Research Design - Formulating The Research Problem

This document discusses research design and methodology. It explains that research design is a structured plan for how to conduct research, while methodology refers to the specific techniques and procedures used. There are three main purposes of research: exploration to develop an initial understanding, description to precisely measure characteristics, and explanation to determine why something occurs. The document outlines different research methods that align with each purpose, such as interviews for exploration and experiments for explanation. It also discusses key concepts like correlation, causation, units of analysis, and threats to validity that must be considered when formulating the research question and choosing a research design.

Uploaded by

Abby Accad
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 15

Research design formulating the research problem

Chapter

4 in Babbie & Mouton (2001)

How to pose proper scientific questions The logic of the research process The research process

O/head

p. 98

Formulating the research question

Research design and research methodology


Design a plan or structured framework of how you intend to conduct the research "A strategic framework for action that serves as a bridge between research questions and the execution or implementation of the research" Methodology refers to the methods, techniques, and procedures that are employed in implementing your research plan (design)

Formulating the research question

Purposes of research
Exploration Description Explanation

Formulating the research question

Exploration
To develop an initial, rough understanding of a phenomenon Methods:

literature

reviews Interviews case studies key informants

Formulating the research question

Description
Precise measurement and reporting of the characteristics of the population or phenomenon What is the case? What is the nature of the relationship? Methods: census, surveys, qualitative studies

NOTE: a sidestep to correlational studies

Formulating the research question

Correlation
See

Chapter 2 in Weiten, for a quick review

Correlation: the degree of relationship between two variables, A and B

Formulating the research question

Direction of correlation:

When A has a high value, B has a high value; when A has a low value, B has a low value = a positive correlation. E.g. the relationship between the amount smoked and the probability of heart disease When A has a high value, B has a low value; when A has a low value, B has a high value = a negative correlation. E.g.: Durkheim: the more socially integrated a society is, the lower the incidence of suicide in that society (p. 23 in Babbie and Mouton). Or: amount of daily exercise and probability of heart disease. Also NO correlation = when two variables do not co-occur (see causation). E.g. studying Psychology and the probability of heart disease.
Formulating the research question 7

Explanation
Why Is x the case? or Is x the relationship? Methods

experimental

NOTE: a sidestep to causation

Formulating the research question

Cause

Three requirements:
Cause

precedes effect A cause co-occurs with the effect The third variable problem

Formulating the research question

Cause and correlation

Cause precedes effect


p.

83, Babbie and Mouton, smoking marijuana and academic performance

Co-occurrence
Correlational

research only tells you two if variables happen together A cause always co-occurs with an effect (Drinking a lot of alcohol) and (feeling lightheaded and throwing up) co-occur because drinking alcohol causes drunkenness
Formulating the research question

10

Cause and correlation

BUT:

non-causally related events can also co-occur!

ALSO: Very few perfect correlations in the probabilistic world of social explanations Two non-causally related events/variables can cooccur because they are both related to something else the third variable

(The ANC wins the 2004 election) and (Dave obtains distinctions in his first semester courses in 2004)

p. 83, Babbie and Mouton, smoking marijuana, academic performance, and emotional problems Does watching a lot of violent programmes on TV, or playing violent video games, cause children to behave more aggressively in preschool?
Formulating the research question 11

Causality:
A

causes B if and only if

exists then B exists A does not exist, then B does not exist Correlational studies test only If A exists then B exists To test for causality, it is generally considered that we need a different type of design: an experimental design. To be discussed under Types of design.
Formulating the research question 12

Some decisions about designs

The purpose of your study (exploration, description, explanation)

This will impact on your decision about the type of design you will need What are you talking about? Individuals? Groups? Organizations? Social actions?

The unit of analysis


The time dimension

Is the study longitudinal (follow people over a long period of time) or cross-sectional (a snapshot in time)?
Formulating the research question 13

Some decisions about designs

Examples

See Weiten p. 434, on the long-term stability of temperament and personality: is a child's temperament at ten a predictor of his/her temperament at ten? Those of you doing Developmental Psychology (PSY209F), look at Chapter 1 in Sigelman & Rider Also distinguish between trend, cohort and panel studies Trend studies: changes in a population over time Cohort studies: changes in relatively specific sub-populations (cohorts) as they change over time Panel studies: examine the same set of people over time.

Formulating the research question

14

Conclusion

These decisions affect the conclusions that can be drawn. Hence careful consideration of them PRIOR to the studys commencement. We want our research to be VALID: truth, rationality, objectivity. A study has validity when it has the capacity to study what it aims to study e.g. one that claims to study intelligence must have some measure of intelligence in it. A study with poor validity is powerless. Different types of research design have different threats to validity.
Formulating the research question 15

You might also like