Research Questions

1
Research Questions Hypotheses

• January 13

2
Overview

• What is a research question?
• How does one develop one?
• How does one evaluate one?

3
Objectives After today you should be able to ...

• Understand the importance of a well-developed
  research question.
• Be aware of numerous methods for generating a
  research question.
• Develop a concise research question.
• Be able to evaluate the quality of a research
  question.
• Understand the role and nature of publishable
  replication research

4
THE IMPORTANCE OF THE RESEARCH QUESTION

• The research question is the starting point of
  the study. Everything flows from the research
  question. It will determine the population to be
  studied, the setting for the study, the data to
  be collected, and the time period for the study.
  A clear and concisely stated research question is
  the most important requirement for a successful
  study.

5
(No Transcript)
6
PREREQUISITES

• The most important prerequisite for this activity
  is a well-cultivated curiosity. This seems to be
  a common characteristic possessed by notable
  researchers. Beyond being curious, these
  individuals also had the patience and tenacity to
  follow a question until satisfied with the answer.

7
Origins of a Research Question

• Careful Observation of People
• Application of New Technology
• The Annoyance Principle
• Build on Experience
• Scientific Communications
• Skeptical Attitude
• Bias (in book) are opportunities
• McGuires Heuristics

8
Some of My Favorites

• Question construct validity
• Confounds, old wine
• Question relationships
• Direction of causality
• Third variables
• Moderators mediator
• Levels of analysis (can increase)
• boundaries (generalizability) - external validity
• need good reason to question
• Settings (e.g., situational specificity
  hypothesis)
• types of individuals/units (children/adults
  e.g., Zollo study)
• Is DV end of the story? (e.g., behavior effects
  setting)

9
Characteristics of a good research question

• FINER
• Feasible
• Adequate numbers of subjects?
• Adequate technical expertise?
• Affordable in time and money?
• Is it possible to measure or manipulate the
  variables?
• Interesting
• To the investigator?
• Novel
• To the field?
• Ethical
• Potential harm to subjects?
• Potential breech of subject confidentiality?
• Relevant
• To scientific knowledge/theory?
• To organizational, health or social management
  and policy?
• To individual welfare?

10
Hypotheses (for Analytical)

• What do you expect to find?
• alternative or experimental hypothesis v.
• null hypothesis
• Statistical v. research
• Examples
• RQ Is a happy worker a productive worker?
• H1 Happier workers are more productive than
  unhappy workers.
• RQ Does increasing the happiness of workers
  make them more productive?
• H1 Increasing the happiness of workers does not
  increase productivity.

11
Should hypotheses be developed before data are
collected or after?
12
Good hypotheses

• Constructs are clear
• Relationship (sign, direction if experimental,
  type of moderation) is clear
• Population often included
• Design/statistical method often clear
• Mean differences
• Compared to who? (cant have a more without a
  than
• Related (correlation)
• The word significant is unnecessary

13
Good hypotheses

• Statistical test is clear (usually one per
  hypothesis)
• With mediator hypothesis may be
• X will positively relate to Y
• M will positively relate to Y
• X will positively relate to M
• X will not relate to Y when controlling for M
• OR
• M will mediate the positive relationship between
  X and Y