Research in Behavioral Science

1
Research in Behavioral Science

• Variable
• Any attribute or property in which organisms vary
• Must consist of at least 2 values

2
Common Goal of Behavioral Research

• Behavioral variability
• Why and how behavior
• varies across situations
• Differs among individuals
• Changes over time

3
Analysis of Research

• Descriptive Statistics
• Summarize and describe behavior of subjects
• Inferential Statistics
• Answer questions
• How likely are findings due to random extraneous
  factors?
• How representative are findings of the larger
  population from which sample came?

4
Variance An Index of Variability

• Variance
• Measure of dispersion
• Range
• Standard Deviation

5
Variance

• Total Variance is sum of
• Systematic variance error variance

6
Variance

• Systematic Variance
• When some of total variability in one variable is
  related in a systematic fashion to variability in
  another variable

7
Variance

• Error variance
• Portion of total variance that is unrelated to
  the variables under investigation in a study

8
Variance

• Distinguishing Systematic vs Error variance
• Proportion of total variance that is systematic
  variance
• Systematic variance / total variance

9
Steps to calculate variance

• Calculate the mean
• Subtract mean from each subjects score to obtain
  the deviation score
• Square each participants score
• Sum the squared deviation scores
• Sum of Squares
• Divide number of scores minus 1
• variance

10
Another goal of behavioral research

• Internal validity
• Contingent on research design

11
Experimental Design

• Purpose of a good experimental design
• to minimize extraneous variation in order to
  increase likelihood that an experiment will
  produce internally valid results

12
Experimental Design

• Between-Subjects Design
• Within-Subjects Design

13
Measurement

• Properties of event or attribute are being
  represented by names or numbers
• Way we assign those numbers/names determines a
  measurement scale
• Determines the conclusions we can draw

14
Scales of measurement

• __________ is the weakest type
• Examples include
• Gender
• Diagnostic category
• Hair color
• Nominal

15
Scales of measurement

• ________ involves the rank ordering on a
  particular dimension, but does not indicate the
  distance between participants on the dimension
• Examples include
• Small, medium, and large
• high, moderate, and low
• Letter grades
• Ordinal

16
Scales of measurement

• _______ is used when equal differences between
  the numbers reflect equal differences between
  participants in the characteristic being
  measured however, the scale does not have an
  absolute zero
• Examples include
• IQ
• Fahrenheit thermometer
• Interval

17
Scales of Measurement

• ________ is the highest level of measurement. It
  has a true zero point, invovling numbers that can
  be added, subtracted, multiplied, and divided
• Examples include
• Weight
• Scores on certain tests
• Ratio