ch6

• Research Strategies


  The general approach and goals of a research study
• Descriptive Research Strategy


  • Examining Individual Variables
  • Describes individual variables (rather than a relationship between variables)
  • Obtains a snapshot (a description) of specific characteristics of a specific group of individuals
  • Data are usually in the form of averages or percentages
• Strategies that Examine Relationships between Variables

  • Most research examines the relationship between variables
  • Changes in one variable are accompanied by changes in another variable
  • Relationships between variables may be: linear, curvilinear, positive, negative
• Different Types of Relationships between Variables

  • A General Relationship
  • Positive Linear
  • Negative Linear
  • Positive Curvilinear
• Correlational Research Strategy


  • Measuring Two Variables for each Individual
  • Consistent patterns are better seen in a graph called a scatter plot
  • Each individual is represented by a point
  • Correlation does not imply causation
  • It describes a relationship but does not explain it
• Comparing Two or More Sets of Scores


  • The experimental, quasi-experimental and non-experimental research strategies
  • Compares two or more groups of scores
  • One of the variables differentiates the groups
  • The second variable measured obtains a score for each individual
• Comparing Grades: High-School Grades for Students from High- and Low-Income Families


  • High Income
  • 72, 86, 81, 78, 85, 80, 91, Mean = 81.9

  Low Income

  • 83, 89, 94, 90, 97, 89, 95, Mean = 91.0
• Experimental Research Strategy


  • Answers cause-and-effect questions about the relationship between two variables
• Quasi-Experimental Research Strategy


  • Almost, but not quite, experiments—can never produce an unambiguous explanation
• Non-Experimental Research Strategy


  • Demonstrates a relationship between variables—does not attempt to explain it
• Data: Experimental, Quasi-Experimental and Non-Experimental

  • Experimental: Low exercise, High exercise

  Quasi-experimental: Without treatment, With treatment

  Non-experimental: Girls, Boys
• Non-Experimental and Correlational Research

  Same goal: Both are designed to demonstrate that a relationship exists between two variables

  Do not try to explain the relationship

  Correlational uses one group of participants, measures two variables for each person

  Non-experimental compares two groups of scores, measures only one variable for each individual
• Research Strategy Summary


  • Descriptive

  Correlational

  Experimental

  Quasi-experimental

  Non-experimental
• Category 1: Descriptive


  • Purpose: produce a description of individual variables as they exist within a specific group

  Data: a list of scores obtained by measuring each individual in the group being studied
• Category 2: Correlational


  • Purpose: produce a description of the relationship between two variables but do not attempt to explain the relationship

  Data: measure two variables (two scores) for each individual in the group being studied
• Category 3: Experimental


  • Purpose: produce a cause-and-effect explanation for the relationship between two variables

  Data: create two treatment conditions by changing the level of one variable; then measure a second variable for the participants in each condition
• Category 3: Quasi-Experimental


  • Purpose: attempt to produce a cause-and-effect explanation but fall short

  Data: Measure before/after scores for one group that receives a treatment and for a different group that does not receive the treatment
• Category 3: Non-Experimental


  • Purpose: produce a description of the relationship between two variables but do not attempt to explain the relationship

  Data: measure scores for two different groups of participants or for one group at two different times
• Research Designs


  • Require decisions about three basic aspects of the research study:

  Group versus individual

  Same individuals versus different individuals

  The number of variables to be included

  Provide a general framework for conducting studies
• Research Procedures


  • Details about how the study is to be done

  Exact, step-by-step description of a specific research study

  Includes a determination of:

  Exactly how the variables will be manipulated, regulated and measured

  Exactly how many individuals will be involved

  Exactly how the individual participants will proceed through the course of the study
• Data Structures and Statistical Analysis (1 of 2)

  • Experimental, quasi-experimental and non-experimental studies:

  All involve comparing groups of scores

  Use similar statistical techniques (e.g., t-tests, analysis of variance and chi-square tests)

  Correlational studies:

  Numerical scores: analyzed with a correlation calculation (e.g., the Pearson correlation)

  Non-numerical data: evaluated using chi-square test
• Data Structures and Statistical Analysis (2 of 2)

  • Descriptive studies summarize single variables for a specific group of individuals:

  Numerical data: analyzed by a statistical calculation of the mean score

  Non-numerical data: evaluated by a report of the percentage associated with each category
• External Validity


  The extent to which the results of a research study can be generalized
• Threat to External Validity

  Any characteristic of a study that limits the ability to generalize the study's results
• Three Different Kinds of Generalization


  • Generalization from a sample to the general population

  Generalization from one research study to another

  Generalization from a research study to a real-world situation
• Internal Validity


  Concerned with factors in the research study that raise doubts or questions about the interpretation of the results
• Threat to Internal Validity


  Any factor that allows an alternative explanation for the results
• Validity and the Quality of a Research Study

  Determined by the extent to which the study satisfies the criteria of internal and external validity

  Threat to validity: Any factor that generates doubts about the accuracy of the results or raises questions about the interpretation of the results
• Research studies vary in terms of validity
• Never accept a research result or conclusion as true simply because it has been 'scientifically demonstrated'
• Being aware of threats to validity can help you critically evaluate a research study
• Make your own decisions about a research report's validity and quality
• Three General Categories of Threats to External Validity

  • Generalizing across participants or subjects

  Generalizing across features of a study

  Generalizing across features of the measures
• Threats to Generalizing across Participants or Subjects

  • Selection bias

  Over-reliance on college students

  Volunteer bias

  Participant characteristics

  Cross-species generalizations
• Threats to Generalizing across Features of a Study

  • Novelty effects

  Multiple treatment interference

  Fatigue

  Practice

  Experimenter characteristics
• Threats to Generalizing across Features of the Measures

  • Sensitization

  Generality across response measures

  Time of measurement
• Extraneous Variables


  Any variables in a research study other than the specific variables being studied
• Confounding Variables


  Extraneous variables (usually unmonitored) that provide an alternative explanation for the observed relationship between the two variables
• Extraneous Variables and Threats to Internal Validity


  • Environmental variables (general threats for all designs)

  Participant variables (individual difference)

  Time-related variables (threats for designs that compare one group over time)
• The goal of any research study is to maximize internal and external validity