Non Experimental Research Designs Survey Research

1
Non Experimental Research Designs Survey
Research

• McBurney White
• Ch.10

2
Survey Research

• Surveys provide information about peoples
  attitudes, thoughts, feelings, behaviors
• Commonly used in correlational research
• Widely used in psychology
• Benefits of surveys
• Easy to administer
• Can collect large amount of data quickly and
  easily
• Useful in generating hypothesis

3
Single-Variable Research

• Potential problems with surveys
• Response rate may be low
• Confidentiality
• Sensitive topics
• Measurement reactivity
• Social desirability
• yea-saying
• Cannot draw causal conclusions from data!

4
How are surveys administered?

• Interviews
• In person or on the phone
• Structured vs. unstructured
• Questionnaires
• Good construction is important!

5
Sampling and Generalization

• Populations vs. Samples
• Importance of obtaining a representative sample

6
Sampling

• How are samples formed?
• Random Sampling
• Simple random sampling
• Stratified random sampling
• Cluster Sampling
• Non-Random Sampling
• Snowball sampling
• Convenience Samples

7
Simple Random Sampling

• Every person in the population has an equal
  chance of being included in the sample.
• Random number tables
• Is it possible to have a true random sample in a
  psychology study?
• Why or why not?

8
Stratified Random Sampling

• In a stratified sample, the population is divided
  into strata (groups) and then members of each
  group are randomly selected
• For example, you might divide all of the
  restaurants in Columbus (the population) into
  groups based on the price ranges of their entrees
  (Strata)
• Under 10
• 10 - 15
• 15 - 20
• Over 20
• Then you could select a sample by randomly
  selecting one restaurant from each category.

9
Stratified Random Sampling

• This method allows you to make sure that your
  final sample contains equal numbers of
  individuals from important subgroups
• Gender age race education level Other
  Examples?
• What if your population is 40 Male and 60
  Female?
• You can design your sample to be 60 female so it
  has the same proportion of males and females
  found in the population
• How does this method impact the generalizability
  of a study?

10
Cluster Sampling

• In a cluster sample, the population is divided
  into groups and then some of the groups are
  selected randomly
• Your population is all college students in the US
• You start with a list of all colleges in the US
  and randomly select 20 schools to be included in
  your sample
• You then randomly choose 50 students from each
  school to be in the final sample

11
Non-Random Sampling

• Snowball Sampling
• Your start with a few individuals in your
  population of interest and then use them as leads
  to other participants
• Ex. A survey of drug use or sexual activity among
  gang members
• Convenience Sampling
• Using whoever is conveniently available for a
  sample
• Most often used method in psychology
• Non-Random sampling limits your ability to
  generalize the results of your study
• These samples are rarely representative
• Can be effectively used to test hypotheses

12
Making sense of your results

• How does the type of sample influence the
  conclusions you can make?
• Generalize only to the population actually
  sampled.
• Most research in psychology uses a convenience
  sample of college students
• Example Gender differences in reactions to
  infidelity
• Would the results differ if you included older
  participants who are married with kids?