Lecture 7, Unit 7: The Summit

1
Exam Study Guide
2
ABA design 

• A baseline period where no treatment is given
  and/or no variable is introduced (A),
• followed by a period in which the treatment or
  variable is introduced (B),
• and then a period in which the treatment is
  removed so the behavior can be observed a second
  time (A).
• This way you can measure behavior before
  treatment, during treatment, and once treatment
  is removed.

3
ABAB design

• The A-B-A-B design represents an attempt to
  measure
• a baseline (the first A),
• a treatment measurement (the first B),
• the withdrawal of treatment (the second A),
• the re-introduction of treatment (the second B).

4
The A-B-A-B design involves two parts

• (1) gathering of baseline information, the
  application of a treatment and measurement of the
  effects this treatment and
• (2) measurement of a return to baseline or what
  happens when the treatment is removed and then
  again applying the treatment and measuring the
  change.

5

• ABAB design has both ethical and internal
  validity advantages over the ABA design.

6
Analytical steps include 

• A)  Apply the statistical test to determine
  whether differences or relationships exist.
• B)  Determine if the difference or relationship
  found is the relationship or difference
  predicted.
• C) Determine the statistical importance of the
  difference or relationship. 
• D)  Generalize the findings.

7
Scholarly Database Research

• AND between concept terms yields fewer results
  than OR between terms. 

8
StatisticsANOVA represents analysis of variance.

• Analysis of variance (ANOVA) is used to uncover
  the main and interaction effects of categorical
  independent variables (called "factors") on an
  interval dependent variable.
• Independent variable is nominal level data
  (categories).
• Two-way ANOVA This statistic can test for an
  interaction effect of more than one independent
  variable.

ANOVA
9
Variance

• In probability theory and statistics, the
  variance of a random variable is a measure of its
  statistical dispersion, indicating how far from
  the expected value its values typically are.
• The F-test is a measure of variance

10
Chi-square

• Computed on categorical or nominal data.

11
Contingency analysis

• Two-way chi-square, two-dimensional chi-square,
  contingency table are names for a contingency
  analysis.

12
t-test

• The t-test tells us if the variation between two
  groups is "significant".
• One-tailed t-test does not test for a significant
  difference in either direction.
• t-test compares means.
• A paired comparison t-test does not look for the
  relationship between two continuous-level
  variables.

13
is the standard deviation
14
Use the Internet or Use the Final Exam Study Guide

• 1. What is ABA design?2. What is ABAB design?
• 3. What are analytical steps in using a
  statistical test?
• 4. What is an ANOVA?5. What is a Chi-square?
• 6. What is a contingency analysis?
• 7. What is a t-test?

Discuss
15
VariablesCombined set of variables 

• A multiple correlation is a relationship between
  one variable and a combined set of variables.

16
Confounded variables muddy the waters

• A researcher studied the effect of defendant
  physical attractiveness on juror decisions.
• The attractive person was 20 years old, and the
  unattractive person was a 45-year old.
• The problem here is that age is confounded with
  attractiveness.

17
Conceptual replication

• A study that uses different procedures for
  manipulating and measuring variables from another
  study is called a conceptual replication. 
• Can you think of an idea from a research study
  you read, which could be translated into a
  different design?

18
Conditions

• A researcher studied the effect of defendant
  gender (male-female) and type of crime
  (robbery-embezzlement) on juror decisions.
• There are four conditions in this study.

19
Main effect

• Effect of one independent variable averaged over
  the other independent variables.
• The main effect is the most powerful.
• Main effect of type of crime  Participants gave
  longer sentences for embezzlement than robbery,
  irrespective of gender. This implies that there
  was a main effect of type of crime.

20
Mortality

• Refers to the fact that participants may drop out
  of experiments.

21
Mundane realism 

• Realistic reconstruction for experiment.
• A study on fear of medical procedures was
  conducted in a room on campus that had been
  decorated as a doctor's office has mundane
  realism.

22
Panel study 

• A panel study is used to study changes in survey
  responses over time.
• Measure the same sample of respondents at
  different points in time.
• Can reveal both net change and gross change in
  the dependent variable.
• Can reveal shifting attitudes and patterns of
  behavior that might go unnoticed with other
  research approaches.

23
Positive linear relationship

• Increases in the levels of one variable are
  associated with increases in the levels of a
  second variable.

24

• 1. What is a multiple correlation?
• 2. What is a confounded variable?
• 3. What is conceptual replication?
• 4. What is meant by a study with four
  conditions?
• 5. What is a main effect?
• 6. What is mortality?
• 7. What is mundane realism?
• 8. What is a panel study?
• 9. What is a positive linear relationship?

• Discuss

25
Range 

• Range is the array of information.
• You expected to find that age is related to more
  conservative political attitudes.
• However, your study found that the two variables
  were not related.
• Before you conclude that the two variables are
  not related, you should examine the range of
  values on your age variable.

26
Reactive Measure 

• The test changes the behavior being measured.
• The course final exam is designed to measure your
  knowledge of course jargon. By taking the test,
  you learn certain information, which affects your
  behavior on the measure.
• Unobtrusive measures reduce the problem of
  reactivity.

27
Repeated measures (within-subjects)

• The same people participate in each condition of
  an experiment.

28
Single-subject designs

• Considered the design of choice when measuring
  behavioral change or when performing behavioral
  modification.
• Rather than comparing groups of subjects, this
  design relies on the comparison of treatment
  effects on a single subject or group of single
  subjects.
• Can be conducted with multiple participants.

29
Structural equation model example 

• A researcher assessed the following variables in
  a sample of teenage males aggressive behavior,
  anger, loneliness, and perceived social support
  from family and friends.
• In addition to examining the correlations among
  these variables, the researcher tested a proposed
  model of how these variables are related to one
  another.

30
Negative case analysis

• Study the exception.
• A researcher tries to explain why a particular
  observation does not fit with a theory that is
  used to explain other observations.

31
Correlation coefficient

• Provides information about the strength of the
  relationship between variables.

32

• 1. What is range?
• What is a reactive measure?
• What are repeated measures?
• What are single-subject designs?
• 5. What is a structural equation model?
• 6. What is a negative case analysis?
• 7. What is a correlation coefficient?

Discuss
33
Cronbach's alpha 

• A tool for assessing the reliability of scales,
• An indicator of reliability based on the
  correlations of each item in a measure with every
  other.

34
Curvilinear relationship 

• Curvilinear relationships are those in which one
  variable is associated with another variable, but
  the relationship is described by a curve rather
  than a straight line.
• You must have at least three levels of the
  independent variable in your experiment.

35
Demand characteristics 

• You are concerned that participants in your study
  may be figuring out exactly what you are
  studying.
• Knowing my influence their behavior.

36
Directional hypothesis
Directional hypothesis

• "Nurses' communication of role expectations to
  parents will increase satisfaction with nursing
  care," is an example of a directional hypothesis.
• A nondirectional hypothesis states that a
  relationship or difference will occur but does
  not state the direction of the relationship or
  difference.

37
Experimental realism 

• The participants in the study became very
  involved in a computer simulation.
• This study has experimental realism.

38
Factorial design

• If a study has two or more independent variables,
  it is called a factorial design.
• An independent variable is a hypothesized cause
  or influence on a dependent variable.
• Time in instruction and pull-out predict student
  success.

39
Cohort effect

• The particular impact of a group bonded by time
  or common life experience.
• People of the same age, for example, may have
  certain commonalities.

40
Interaction effect

• The differing effect of one independent variable
  on the dependent variable, depending on the
  particular level of another independent variable.
  
• Example Dr. Fox studied the effect of
  attractiveness on juror decisions. Dr. Ramirez
  studied the effect of type of crime on juror
  decisions. To determine interaction effect, both
  independent variables could be studied in one
  experiment.

41

• What is a Cronbach's alpha?
• What is a curvilinear relationship?
• What are demand characteristics?
• What is a directional hypothesis?
• What is experimental realism?
• What is factorial design?
• What is a cohort effect?
• What is interaction effect?

• Discuss

42
Protection

• Anonymity and confidentiality are two types of
  protection given to participants. 
• Protection for the research team is not
  paramount. 
• Protection given to participants is crucial.
• Beneficence is achieved when the well-being of
  research participants is protected.

43
Ceiling effect

• When a task is so easy that the experimental
  manipulation shows little/no effect.
• Running a pilot study is a good way to determine
  whether a ceiling effect is present. The
  dependent variable was score on a measure of
  comprehension of material.
• Virtually everyone in all conditions of the
  experiment answered all questions correctly.
• Researchers refer to this situation as a ceiling
  effect.

44
Avoid Deception

• Behavioral studies must be reviewed for possible
  sources of physical and psychological stress.
• Audiotaping and videotaping can be used in data
  collection if participants know what is to be
  recorded. 
• Participants' knowing what is to be recorded is
  one agreement that must be secured.

45
Anxiety

• The galvanic skin response (GSR) is a measure of
  anxiety.
• Measures are used in public speaking apprehension
  and deceptive communication studies, for example.

46

• Discuss anonymity and confidentiality.
• What is a ceiling effect?
• What is necessary regarding deception of
  subjects?
• What is galvanic skin response?

Discuss