Untitled Quiz

Experimental Design (2)

• Between-subjects design (independent design) involves participants in different experimental conditions being independent of one another.
• A one-way design has one independent variable (IV), while a factorial design has more than one IV.
• Within-subjects (repeated-measures) design requires participants to undergo all experimental conditions in a single study.
• Random assignment is a method for placing participants equally in different conditions.
• Random assignment is not the same as random sampling. Random sampling aims to match the sample of participants to the wider population.
• Pre-tests can be used to ensure both groups were comparable before the experimental conditions start (e.g., equivalent memory tasks).
• One-way design involves one IV with two or more levels.
• Example IV: use of highlight pens (highlight vs. no highlight) measured against memory.
• Example IV: Money and happiness (no cash, some cash, a lot of cash) measured against mood assessment.
• One-way designs investigate the effect of an IV on a dependent variable (DV).
• A between-subjects factor does not require random assignment. For example, the effect of an academic major on GPA.
• Factorial design in experiments includes two or more IVs.
• Case 1 example of factorial design: visual imagery vs. rote repetition and presenting a word list at different rates (2-sec vs 4-sec) affecting memory recall.

Between-Subjects Factorial Design

• Any study with greater than one IV, each with at least two levels.
• Case 1: Example of a 2x2 design - visual imagery technique vs. rote repetition and visual imagery technique(experimental group) vs. rote repetition(control group) to memorise numbers, for investigating the effect of how quickly words are presented (rate of presentation: 2 secs/word vs. 4 secs/word) on memory.
• Case 2: Example of a 2x2 design involving major in psychology and science as factor 1 (IV #1) and course mode (online vs. face to face) as factor 2 (IV #2).
• This 2x2 design has two main effects and one interaction effect: main effect of major, main effect of course mode and major X course mode interaction effect.

Interaction Effect

• The distinct advantage of factorials over single-factor designs is the opportunity to demonstrate interactive effects.
• An interaction is said to occur in a factorial experiment when one IV's effect depends on the level of another IV.
• In the example, the effect of teaching mode on GPA depends on the student's major.
• For scientific majors, there is no difference between online or face-to-face learning styles. However, for psychology students, face-to-face instruction appears superior to online instruction.
• The effect of variables can vary across conditions.
• Interpretation of a factorial design involves examining the means of the individual cells.
• Order effects occur when participants' reaction to one condition is altered by the prior conditions they have experienced.
• An example of the use of complete/full counterbalancing in an experiment with 2 conditions.

Complete/Full Counterbalancing

• Participants are divided into subgroups of equal size to receive similar treatments, each in a different sequence.
• The example of a study of how a participant's taste (preference) of chocolate is influenced by (comparison) between shared and unshared experience.
• Order effects, such as practice, fatigue, and boredom, become factors in their own right when counterbalancing, reducing variability among the resulting groups.

Incomplete Counterbalancing

• Each treatment appears equally often in each position of the experiment, for example, in the 1st, 2nd, 3rd, and 4th positions.
• A Latin square is used to establish a sequence for repeated measure designs when the number of conditions is too large for complete counterbalancing.
• The first row is generated by a formula, and subsequent rows follow a numerical order down each column.

Mixed Design

• A more complex and advanced experiment design combining between-subjects and within-subjects elements.
• The design always requires at least two independent variables, one using between groups and the other repeated measures.
• An example is the relationship between age and cell phone use.
• Age is used as a between-subjects factor (different participants representing different age groups).
• Cell phone usage (yes/no) is used as a within-subjects factor, requiring each participant to experience both conditions (e.g., talking on a cell phone vs. not talking on a cell phone).

Steps in Experimental Design

• Clear and justifiable motivation of study — identify research gap and justify the motivation.
• Define variables (IVs and DV) and operational definitions.
• Ensure design, procedures, and sample characteristics are detailed.
• Determine the sample size and number of trials.

Conclusion

• Experimental designs such as between-subject, within-subject, mixed, one-way, and factorial designs.
• Participants in each group are carefully selected to be as similar as possible.
• Counterbalancing techniques are used to counter order effects in within-subject designs.