• Does the measure cover the range of meanings of dimensions included within the concept?
• Includes:
• Face Validity
• Expert Panel Validity

• Does a given measure "on it's face" represent the conceptualization of the variable

• Using expert opinion to decide if the measure represents the conceptualization of the variable
• Experts might
• Evaluate the measure post-construction
• Help create the measure

• Predictive Validity
• Concurrent Validity

• Does the measure predict future behavior?

• Do participants scores on the measure reflect scores on similar measures

• Convergent Construct Validity
• Discriminant Construct Validity

• Test-Retest
• Internal Consistency
• Split-half reliability
• Item-total reliability
• Reliability Coefficient

What is Split-half reliability?

(Part of Internal Consistency Reliability)

• Randomly choose two subsets of items
• Subset should be highly correlated

What is Item-total reliability?

(Part of Internal Consistency Reliability)

• Scores on individual times should positively correlate with the total scores on the scale

• Can capture the correlation between items with a reliability coefficient
• Most common is Cronbach's alpha
• Ranges from 0.00 - 1.00
• Researchers hope to produce values close to 1.00
• General cutoff for reliability is .70

True Experimental Design

(Experimental Designs)

• True Experiments involve
• A Control Group
• An Experimental Group
• Random Assignment to these groups

• X
• Introduction of the Independent Variable (IV)
• The Experimental Treatment
• O
• Observation of a Dependent Variable (DV)
• R
• Random Assignment

Pretest-Posttest Control Group Design

(Experimental Designs)

R    O1        X      O2 

R    O3                O4

Posttest Only Control Group Design

(Experimental Designs)

R           X        O1 

R                      O2

Solomon Four-Group Design

(Experimental Designs)

R    O1        X      O2 

R    O3                 O4

R                  X       O5 

R                           O6

• Experiments that lack full experimental control
• Typically involve some type of comparison group
• But the comparison group is not randomly assigned

The Time Series Design

(Quasi-Experimental Designs)

 O1 O2 O3 O4 X O5 O6 O7 O8

Nonequivalent Control Group Design

(Quasi-Experimental Designs)

O1   X  O2

O3    O4

Multiple Time Series Design

(Quasi-Experimental Designs)

O1 O2 O3X O4 O5 O6

O7 O8 O9     O10 O11 O12

• The Hawthorne Effect/The Placebo Effect
• Maturation
• Experimental Mortality

• Threat to validity that occurs when participants are not selected randomly
• Happens when intact groups are used to conduct studies
• Differences found at the end of the study may be the result of initial differences between the groups and not the program being studied

A threat to internal validity from an reduction in effectiveness of study participants

• If a participant drops out or dies that was randomly assigned it changes the validity of the study

When participants' attitudees towards being involved in a study affect the way they behave

(Named after the study where they changed the brightness of the lights in the work room, done at Hawthorne Plant of Western Electric Company in Chicago)

• Quantitative Research
• Reality is Objective
• Generalizable explanations of behaviors
• Fixed hypotheses
• Qualitative Research
• Reality is Subjective
• Unique explanations of behaviors
• Evolving hypotheses

• Simply a generalization about a phenomenon
• An explanation of how or why something occurs

• Something that varies, is manipulated, controlled and measured
• Anything that has two or more values to it, & opposite a constant

• Open-ended questions:  Questions in which the respondent is asked to provide his or her own answers
• Gives way for bias and misunderstanding on the part of the researcher
• Closed-ended questions:  Survey questions in which the respondent is asked to select an answer from a list provided by the researcher
• Greater uniformity of responses and more easily processed than open-ended questions

• Contingency
• "Skip 2" questions or referral questions
• Matrix
• Categories range from "strongly disagree" to "strongly agree"