Research Issues & Controls (College Board AP® Psychology): Study Guide

Random assignment

• Psychological research must control for factors that could distort results

  • If these factors are not controlled, researchers cannot confidently conclude that the IV caused changes in the DV

  • Strong research design protects internal validity and reduces bias

• Random assignment is the process of placing participants into experimental or control groups by chance

  • Its purpose is to:

    • distribute participant differences evenly across groups

    • reduce confounding variables

    • strengthen causal conclusions

• Random assignment helps ensure that pre-existing differences (e.g., motivation, intelligence, personality) are unlikely to systematically affect one group more than another

  • These pre-existing differences are a key source of confounding variables — variables other than the IV that could explain changes in the DV

Experimental vs. control group

• Experiments usually consist of two conditions where the effect may need to be compared:

  • The experimental group receives the treatment

  • The control group does not receive the treatment, therefore acts as a baseline for comparison

• Without a control group, it is difficult to determine whether changes in the DV were caused by the IV

• Without random assignment, even with a control group, pre-existing differences between the groups could act as confounding variables and undermine the validity of the findings

Placebo method

• A placebo is an inactive substance or fake treatment given to the control group

• Example:

  • Experimental group receives a new anxiety-reducing drug

  • Control group receives a sugar pill (the placebo)

• Crucially, the control group does not know they have received a sugar pill rather than the real drug

  • Before the study begins, all participants are informed that they may receive either the real treatment or a placebo, and they consent to this as part of the study

  • This means participants in both groups believe they may be receiving the real treatment, which is what allows the placebo effect to be properly controlled

• The placebo controls for the placebo effect — when participants improve simply because they expect to improve

  • If both groups show similar improvement, the results may be due to expectation rather than the treatment itself

  • The placebo method strengthens internal validity by isolating the true effect of the IV

Experimenter bias

• Experimenter bias occurs when a researcher's expectations influence the outcome of a study

• This can happen if the researcher:

  • unintentionally gives cues about the expected outcome

  • treats one group differently to the other group

  • interprets ambiguous results in a biased way

• Because researchers often know the hypothesis, this bias may be unconscious

  • For example, if a researcher has hypothesised that the experimental group will score higher, they may unconsciously communicate this expectation to participants through their tone of voice, body language, or the way they deliver instructions

  • This can influence how participants behave, distorting the results in the direction the researcher expected

Controlling experimenter bias: double-blind procedure

• One way to control for experimenter bias is to use a double-blind procedure

• In a double-blind study:

  • Participants do not know which condition they are in

  • The researcher interacting with participants also does not know

  • This prevents both participant and researcher expectations from influencing results

Demand characteristics and social desirability bias

• Demand characteristics occur when participants pick up cues about the purpose of a study and alter their behaviour accordingly

• Participants may:

  • try to confirm what they think the hypothesis is

  • try to sabotage the study

  • act how they think they are supposed to act

• This makes results artificial and threatens the internal validity of the study

• Social desirability bias is a specific type of demand characteristic — where participants respond in ways they believe are socially acceptable rather than behaving truthfully

  • This is especially common in surveys, interviews, and studies on sensitive topics (e.g., prejudice, substance use, aggression)

  • For example, participants may underreport aggressive behavior or overreport charitable acts to appear more favorable

Controlling demand characteristics: single-blind procedure

• One way to control for demand characteristics is to use a single-blind procedure

• In a single-blind study:

  • participants do not know which condition they are in

  • this reduces the likelihood that they will modify behavior based on what they think is expected of them

• If participants cannot identify which group they are in, they are less able to respond according to perceived demand characteristics or social pressure