Experimental Psychology, 7th by Anne Myers solution Manual + Test Bank

Experimental Psychology, 7th by Anne Myers solution Manual + Test Bank

Chapter 7

The Basics of Experimentation

Chapter Overview

Independent and dependent variables are central to experimental research. The independent variable is the antecedent condition an experimenter deliberately manipulates to assess its effect on behavior. There are three types of independent variables (environmental, task, and subject). The dependent variable indicates change in behavior. Researchers create operational definitions for independent (experimental operational definition) and dependent (measured operational definition) variables. Operational definitions provide precise meanings for both unseen processes (hypothetical constructs) and observable behavior (nonconstruct variables) in an experiment, and allow replication by others.

In designing an experiment, researchers can choose among four levels of measurement: nominal, ordinal, interval, and ratio. These scales provide increasing information about a variable. Nominal scales completely ignore magnitude and assign items to distinct categories on the basis of a common feature. Ordinal scales rank items on the basis of magnitude. Interval scales measure magnitude along scales with equal intervals between all values. Ratio scales measure magnitude along scales that have both equal intervals between values and a true zero point. When several levels of measurement “fit” equally well, the authors recommend selection of the highest level because it provides the most complete information about a variable.

Researchers evaluate operational definitions in terms of reliability and validity. Reliable procedures produce consistent and dependable outcomes. Three methods of checking the reliability of measurement techniques include interrater reliability, test-retest reliability, and interitem reliability. Valid operational definitions manipulate and measure the variables we intend to study. Five criteria for evaluating validity include face validity, content validity, predictive validity, concurrent validity, and construct validity.

An experiment is internally valid if we can be sure that the changes in behavior that occurred across treatment conditions were caused by the independent variable. Sometimes we discover that extraneous variables, variables that are neither independent nor dependent variables, also change across conditions. When both extraneous and independent variables change value from one condition to another, we have a situation known as confounding. Confounding threatens internal validity because it prevents us from inferring a causal relationship between the independent and dependent variables. Campbell and colleagues have identified eight classic threats to internal validity. These threats include history, maturation, testing, instrumentation, statistical regression, selection, subject mortality, and the set of selection interactions.

The chapter concludes with an examination of the purpose and content of the Method section of the research report, and planning suggestions.

Chapter Outline

Independent and Dependent Variables
Some Research Examples
Identifying Variables
Operational Definitions
Defining the Independent Variable: Experimental Operational Definitions
Defining the Dependent Variable: Measured Operational Definitions
Defining Constructs Operationally
Defining Nonconstruct Variables
Defining Scales of Measurement
Evaluating Operational Definitions
Reliability
Validity
Evaluating the Experiment: Internal Validity
Extraneous Variables and Confounding
Classic Threats to Internal Validity
Planning the Method Section
Summary
Key Terms
Review and Study Questions
Critical Thinking Exercise
Online Resources

Key Terms

Concurrent validity Levels of the independent variable
Confounding Manipulation check
Construct validity Maturation threat
Content validity Measured operational definition
Dependent variable (DV) Method
Experimental operational definition Operational definition
Extraneous variable Predictive validity
Face validity Reliability
History threat Selection interactions
Hypothetical construct Selection threat
Independent variable (IV) Statistical regression threat
Instrumentation threat Subject mortality threat
Interitem reliability Testing threat
Internal validity Test-retest reliability
Interrater reliability Validity
Level of measurement

Teaching Suggestions

Cengage Online Workshop Exercises

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/confounds/confounds_01.html
One of the Research Methods workshops is titled “Confounds – Threats to Validity.” Have students work through it and then ask them the following:
What good is internal validity in an experiment?
How can you tell if you have construct validity?
What are confounds and why are they bad?
What is differential mortality?
What problems are reduced when you make your study double-blind?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/reli_vali/reli_vali_01.html
This workshop covers both “Reliability and Validity.” Ask your students:
How is reliability different from validity?
How do you know if a survey instrument has internal consistency?
Name and describe three types of reliability.
What is Cronbach’s alpha used to determine?
Which type of validity assesses the match between your measure and the underlying concept?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/controls/controls_01.html
This is a more detailed workshop on various issues of “Control” in experiments. Students should learn to answer:
Why is control an important part of experimentation?
What are inclusion and exclusion criteria?
Define the term “double-blind” and explain its usefulness
How can we control a between-subjects design? A within-subjects design?
What are order effects? How can we control for them?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/mistake/mistake_01.html
The classic threats to internal validity described in the text have a nice overlap with parts of this “Common Mistakes in Student Research” workshop. After reading it, students should be able to discuss these:
What common mistakes occur during the literature review process?
What should you do (and not do) when you discover mixed results in published studies?
Describe some mistakes you can make when selecting research volunteers.
Give three benefits of conducting a pilot study.
What are two reasons why your results might reveal no statistically significant differences?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/manip_chck/manip_chck_01.html
Also relevant to this crucial chapter in the textbook is this workshop on “Manipulation Checks in Experimental Research.” Students can learn answers to these:
What are the two defining features of true experiments?
What is the purpose of a manipulation check in an experiment?
Describe both an obvious and an unobtrusive way to check your manipulation.
What are some of the manipulation checking techniques used in clinical studies that involve medications?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/op_def/op_def_01.html
This workshop titled “Operational Definitions” has a section on experiments that discusses reliability and validity. Ask your students the following after they’ve studied this workshop:
Why are operational definitions important?
What considerations are important when using behavioral observation to collect data?
Is pilot testing needed when behavioral observation is being used? Why/why not?
What issues should be considered when using surveys to collect data?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/stat_workshp/scale/scale_01.html
In the Statistics Workshops section of this website is a unit on “Scale of Measurement” that does a good job covering this important element of experiment construction. Students should learn the following:
What are the four scales of measurement, in order?
How does an ordinal scale differ from a nominal scale?
In which scale are the assigned values arbitrary? Explain.
What three properties are shared by all interval scales?
What types of comparisons are only possible with ratio scales?
What fifth scale of measurement is suggested in this workshop?

http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/res_methd/constrcts/constrcts_01.html
Lastly, there is a research methods workshop about “Specifying Constructs” that discusses the difficulties related to variables that cannot be directly observed. Students who study this will learn answers to the following questions:
What is a conceptual definition? Give a conceptual definition of “sleepy.”
Describe the challenges involved in operationally defining PTSD.
What should you do if your construct has multiple dimensions?

Additional Web Resources

http://www.socialresearchmethods.net/kb/index.php
This Research Methods Knowledge Base from William Trochim at Cornell has some very good content about internal validity, sampling, measurement, and other experimental issues.

http://www.apastyle.org/
This site is part of the web resources of the American Psychological Association. It provides information about APA format and publications on the topic.

http://onlinestatbook.com/rvls.html
http://www.ruf.rice.edu/~lane/stat_sim/reg_to_mean/
Rice offers a “Virtual Lab in Statistics” which includes a page on regression toward the mean (one of the classic threats described in the textbook).

Classroom/Homework Exercises

Handout 7-1: Conceptual and Operational Definitions

For each of the terms listed below provide a conceptual definition (a commonly accepted description of the term), an experimental operational definition (as if the term represented an IV manipulated in an experiment), and a measured operational definition (as if the term represented a DV measured in your experiment).

Intoxicated
Conceptual:

 

Experimental Operational:

 

Measured Operational:

 

Religious
Conceptual:

 

Experimental Operational:

 

Measured Operational:

 

Introverted
Conceptual:

 

Experimental Operational:

 

Measured Operational:

 

Optimistic
Conceptual:

 

Experimental Operational:

 

Measured Operational:

 

Immature
Conceptual:

 

Experimental Operational:

 

Measured Operational:

Levels of Measurement
This chapter describes four levels of measurement a variable might have (nominal, ordinal, interval, ratio). With a little creativity, many variables can be found to work with two, three, or even all four levels. For each of the following terms, ask the class to indicate all levels that are possible for the variable (each has more than one). Have them explain their answers. Alternatively, you can also break your class into small groups for this exercise and have the groups compete to identify the most levels.

Intelligence

 

The “hotness” of chili in a bake-off contest

 

Memory

 

Tendency to procrastinate

 

Membership in student organizations