Research Methods and Ethics Quiz

Experimental Design I: Single-Factor Designs

• Single-factor designs have one independent variable.
• Four varieties of single-factor designs: independent groups, matched groups, nonequivalent groups, and repeated measures.
• Independent groups designs use different groups for each level of the independent variable.
• Matched groups designs try to form equivalent groups by matching characteristics.
• Nonequivalent groups involve groups that are not equivalent. This can be a limitation.
• Repeated measures designs involve the same participants at each level of the independent variable.
• Bar graphs and line graphs are used to present data. Bar graphs are used to show the differences between groups, and line graphs show within-subject data over time.
• Types of control groups used: placebo, wait-list, and yoked control groups.
• The one-way ANOVA is used if you are examining data from an independent-group or multilevel study instead of multiple t-tests. This reduces the possibilities of errors in analysis.
• Post-hoc analyses are commonly used with one-factor ANOVAs. This post-hoc testing is used to ensure all possible comparisons are made between the levels of the independent variable once an overall significant effect has been found.

Ebbinghaus Memory Research

• Ebbinghaus aimed to understand memory processes.
• His methodology focused on self-testing memory.
• Results showed a fast initial rate of forgetting, followed by a slower decline.

Single-Factor-Two Levels

• Between-subjects, single factor designs examine independent groups.
• Manipulated independent variables are deliberately varied to see how they affect the dependent variables.
• Random assignment ensures equivalent groups.
• Examples provided include note-taking methods and social skills training.
• Subject variables are intrinsic characteristics that cannot be manipulated. These may cause nonequivalence between groups.

Ex Post Facto Designs

• Subject variables are used as independent variables.
• Researchers consider how an existing condition impacts a dependent variable without manipulating it.
• Attempts include carefully selecting participants to minimize nonequivalence between groups.
• Examples include studies of traumatic brain injury effects.

Within-subjects, Single-Factor Designs

• Also referred to as repeated-measures designs.
• Participants are included in all levels of the independent variable.
• Often use counterbalancing techniques.
• Examples are the Stroop effect and chocolate preference.

Single-Factor-More Than Two Levels

• Between-subjects, multilevel designs can show non-linear effects.
• Examples include the relationship between arousal and performance, and the impact of contextual information on memory.
• A significant advantage of the multilevel design is to ensure discovering non-linear effects.

Analyzing Data from Single-Factor Designs

• Different graphs are used to represent data, depending on if comparing groups (bar) or within-subject changes over time (line).
• Table 7.2 displays a comparison of recall data.
• Ebbinghaus forgetting curve shows retention over time.

Special-Purpose Control Group Designs

• Placebo control groups are used to isolate the effects of a treatment.
• These groups receive an inactive substance or a control treatment.
• Waiting-list control groups ensure equivalent groups.
• Research Example 17 shows an application of these methods in a study of subliminal weight loss tapes.
• A significant aspect of special-purpose methodologies is ensuring a double-blinding procedure.

Yoked Control Groups

• Each subject in the control group is matched with those in the experimental.
• Examples of use in stress study with EMDR therapy.
• The aim is to account for external variables that could affect the result.

Factorial Designs: Chapter 8

• These involve two or more independent variables.
• Use a standardized notation system (e.g., 2x2, 3x5).
• Data is placed in matrices, and row and column means are calculated.
• Main effects describe the effect of each independent variable.
• Interaction effects show when effect of one independent variable depends on the value of another.
• This chapter includes famous studies (e.g., Jenkins and Dallenbach) as examples.
• It considers different types of factorial designs such as P x E designs. Mixed factorial designs include at least one between-subjects factor and at least one within-subjects factor. A critical aspect of factorial design is counterbalancing.
• The number of participants needed for the study is dependent on whether the factors are between-subjects or within-subjects

Non-Experimental Designs: Chapter 9

• Surveys seek information on attitudes, beliefs, and behaviors.
• Survey design factors: sampling issues, principles, wording, data interpretation, and collection methods.
• Four methods for collecting survey data are examined.
• Correlation analysis examines relationships between variables without inferring cause-and-effect.
• Correlation coefficients (positive, negative, no correlation) are used for the analysis.
• Scatterplots illustrate bivariate relationships graphically.
• Regression analysis predicts variables' scores.
• This chapter discusses interpreting correlational data, including problems like directionality and third variables and solutions such as, cross-lagged panel correlation, and partial correlation.