Experimental Methods - Independent Groups Design PDF

EXPER IMENTAL METHODS- INDEPENDENT GROUP S DESIGN Asst. Prof. Elif Manuoğlu WHY PSYCHOLOGIST CONDUCT EXPERIMENTS Why do you think so? WHY PSYCHOLOG IST CONDUCT EXPER IMENTS Testing the hypotheses derived from theories Solve society’s problems by providing vital information about the effec...

EXPER IMENTAL METHODS- INDEPENDENT GROUP S DESIGN Asst. Prof. Elif Manuoğlu WHY PSYCHOLOGIST CONDUCT EXPERIMENTS Why do you think so? WHY PSYCHOLOG IST CONDUCT EXPER IMENTS Testing the hypotheses derived from theories Solve society’s problems by providing vital information about the effectiveness of treatments in a wide variety of areas Theory development/contribution to a theory Ex: testing the propositions of inhibition theory (personal emotional events vs. superficial topics) Ps writing about emotional topics had better health outcomes than Ps writing about superficial topics If the results of the experiment are consistent with what is predicted by the hypothesis, then the theory receives support WHY PSYCHOLOG IST CONDUCT EXPER IMENTS If the results of the experiment are not consistent with what is predicted by the hypothesis, what happens to theory? Theory may need to be modified and a new hypothesis developed and tested in another experiment TWO FEATURES OF AN EXPERIMENT 1. Manipulation of the Independent Variable Quasi IV “cause” DV Experiment- Can This “cause” elicits the response (DV) typically (Random Sampling) manipulate the IV, but cannot randomly assign participants to 2. Control of Extraneous Variables conditions Can hold variable(s) constant Limit variable to a specific category (Random Assignment) LOGIC OF EXPERIMENTAL RESEARCH An independent variable must have at least two levels (also called conditions). One level “treatment” condition Second level the control (or comparison) condition Often, more than two levels are used for additional comparisons between groups Ex: the effect of different types of photo sharing behavior on Instagram on positive emotions Conditions of  IV: sharing photo IV: 1) Selfie  DV: positive emotions 2) Scene or view 3) Food 4) Social LOGIC OF EXPERIMENTAL RESEARCH  IV: sharing photo Conditions of What about IV: control 1) Selfie condition in 2) Scene or this design?  DV: positive emotions view 3) Food 4) Social FOOD FOR THOUGHT Define what a control condition is, explain its purpose in research on treatment effectiveness LOGIC OF EXPERIMENTAL RESEARCH Manipulate an independent variable in an experiment to observe the effect on behavior, as assessed by the dependent variable Experimental control is gained through manipulation, holding conditions constant, and balancing Internal validity when the three conditions are met: covariation, time-order relationship, and elimination of plausible alternative causes Internal validity: The extent to which a research study accurately tests the specific cause- and-effect relationship it intends to examine. Degree to which a study provides convincing and credible evidence that the changes in the IV caused the observed effects on the DV, without interference from other factors. INDEPENDENT GROUPS DESIGN(BETWEEN- SUBJECTS DESIGN Each group of subjects participates in a different condition of the independent variable Uses random assignment to ensure we have comparable groups The groups are formed so as to be similar on all important characteristics The groups are treated the same except for the level of the independent variable RANDOM SAMPLING VS. RANDOM ASSIGNMENT Population vs. Sample Census: Asking everyone Population Sample Participant Goal of Sampling = Represent the population you want to draw conclusions about BODY IMAGE STUDY Hypothesis: Exposure to very thin body images causes young girls to experience negative feelings about their own body The three conditions of the  Conditions of experiment independent variable 1- Barbie photos allowed researchers to make comparisons 2- Emma photos If they tested only the Barbie condition, it would Control condition: Neutral images related to story be impossible to determine The neutral-image whether those images condition created a influenced girls’ body comparison point dissatisfaction in any way  BODY IMAGE STUDY Result: Young girls exposed to the images of Barbie were more dissatisfied with their body shape than were girls who were exposed to the Emme images or to the neutral images Girls in the barbie condition > girls in the Emma condition = girls in the neutral condition Please evaluate this study in terms of internal validity. Hint: covariation, time-order relationship, and elimination of plausible alternative causes BODY IMAGE STUDY Several factors that could have affected the girls’ attitudes toward their body were kept the same across the three conditions of the experiment, such as They all heard the same story, They looked at the pictures in same amount of time They all received the same instructions and went through same procedures during the expr. All in all, conditions were held constant except for the manipulation. Why is that so? What is the purpose? CONFOUNDING VARIABLE Confounding variable: An extraneous variable that interferes with the ability to draw a clear cause-and-effect relationship between the IV and DV. In other words, it's a variable that is not the focus of the study but can have an impact on the results, potentially leading to incorrect conclusions Holding conditions constant is a control technique to avoid confounding If girls read a different story about Barbie and Emma, this would be a confound confounding internal validity BAL ANCING The importance of forming similar groups in terms of ind. diff. variables The participants in each group should be similar in terms of various characteristics such as their personality, intelligence, and so forth (also known as individual differences ) Individual difference variables refer to the unique characteristics, traits, or attributes that vary from one individual to another within a sample or population (personality, attitudes, intelligence, motivation, demographic characteristics and so on) These variables capture the ways in which people differ from each other, and they can be critical in understanding and explaining various psychological phenomena We establish equivalent groups of participants by balancing, or averaging, individual differences That is why, we use random assignment to balance ind. diff. across the conditions FOOD FOR THOUGHT What happens if Ps in one group are systematically different form the Ps in another group? Consider an individual difference variable that might influence the results of the Barbie study? (confound the results) Whenever possible, the conditions should be held constant If not, they should be balanced THREATS TO INTERNAL VALIDITY Threats are potential alternative explanations for the findings Threats to internal validity should be controlled to make a clear cause-and- effect inference about the effect an IV on DV Problems in experimental research that can result in threats to internal validity  Testing intact groups  Balancing extraneous variables  Subject loss  Demand characteristics and experimental effects THREATS TO INTERNAL VALIDITY Testing intact (nonequivalent) groups  Occurs when researchers want to draw causal inferences about the effects of an IV on a DV in a non-experimental or quasi-experimental setting, where random assignment to groups is not possible  The source of the confounding due to noncomparable groups arises when individuals differ systematically across the intact groups  Ex: administering a new teaching method in high-school students to see the effect of new method on learning performance  Group A: receives the new method  Group B (control group): received the regular method as they used to get  What is the problem here? THREATS TO INTERNAL VALIDITY Balancing extraneous variables  Extraneous variable: Potential variables that are not directly of interest to the researcher but maybe sources of confounding  Minimize the potential influence of extraneous variables and maintain the internal validity of the experiment  A class of procedures to ensure the comparison between groups or conditions more accurate  Allows researchers to draw more reliable conclusions about the effects of the IV on DV  Such as: counterbalancing the order of the treatment/manipulation, experimental design (testing all groups in a standardized way), random assignment and forming equivalent groups THREATS TO INTERNAL VALIDITY Subject loss  When subjects begin an experiment but fail to complete it successfully, the internal validity of the experiment can be threatened  It is important to determine the cause of Ps loss to understand whether the loss is related to any characteristics of Ps (systematic differences btw the Ps who successfully completed the experiment and who didn’t)  Mechanical subject loss: when a subject fails to complete the experiment because of an equipment failure  We can rule out the possibility of systematic differences btw the Ps THREATS TO INTERNAL VALIDITY Selective subject loss: Ps are more likely to drop out of a study or be lost to follow-up in one condition or group compared to others. This can introduce a bias in the sample, potentially affecting the validity of the study's results The initial sample and remaining sample may not be comparable or The exp. and control conditions may not be comparable In turn, validity is threatened To prevent it: pretesting and screening Ps who are likely to drop-out and excluding them T H R E AT S T O I N T E R N A L VA L I D I T Y The effectiveness of a 1-month fitness program on fitness. Eighty people volunteer for the experiment, and they randomly assign 40 to each of two groups Experimental group (40 Ps) Control group (40 Ps) 25 finished 38 finished Do you think that exp. and control groups are comparable? Why or why not? THREATS TO INTERNAL VALIDITY The selective loss of participants in the experimental group likely destroyed the comparability of the groups that was ensured by random assignment The loss is likely due to a characteristic of the participants—their original level of fitness-an individual difference variable This characteristic is relevant to the outcome of the study THREATS TO INTERNAL VALIDITY Demand characteristics and experimental effects  Cues, subtle hints and other information that participants use to guide their behavior in a psychological study  Lead participants to form expectations about the study's purpose or what is expected of them  Influence participant behavior, potentially leading to changes in their responses that are not a genuine reflection of their thoughts, feelings, or behaviors  Ex: Being in the treatment condition may cause Ps to behave in a certain way THREATS TO INTERNAL VALIDITY Experimenter effect: The expectations of the experimenter form Ps If experimenter behave differently to Ps in the different conditions of the experiment For an experimenter, there may be certain expectation from Ps who are in the treatment condition Experimenter may make biased observation and report the observation of Ps in the treatment group in a biased way (in a way to confirm the hypothesis of the study) THREATS TO INTERNAL VALIDITY How to minimize the impact of demand characteristics and experimenter bias? Employing double-blind procedures (where neither the participant nor the experimenter knows the true purpose of the study) Employing deception when necessary to prevent participants from guessing the study's true purpose Using place control group Placebo: Substance that looks like a drug or other active substance but is actually an inactive substance The placebo effect occurs when a person gets better by thinking that a fake medicine that has no pharmacological effect is effective Researchers test the effectiveness of a treatment by comparing it to a placebo 11/15/2024 S a m p l e Fo o t e r Te x t 29  Double-blind procedures: THREATS TO  Neither the participants nor the researchers INTERNAL involved in a study know which Ps are VALIDITY receiving the treatment or intervention and which are in the control group  Purpose: To eliminate biases (demand characteristics and experimenter effect) and control for the placebo effect  To ensure that the observed effect was caused solely by the treatment/manipulation but not by the expectations of the Ps or the biased approach of the experimenter FOOD FOR THOUGHT What can be the potential ethical problems of involving a placebo control group in the experiments? Informed consent, consent of Ps to participate in the study, offering treatment to Ps in the placebo condition after the exp. is over E S TA B L I S H I N G T H E E XT E R N A L VA L I D I T Y O F E X P E R I M E N TA L F I N D I N G S Extent to which findings from a research study can be generalized to individuals, settings, and conditions beyond the scope of the specific study A criticism to lab experiments: lack of external validity , controlled conditions, specific setting with specific Ps When the purpose of an experiment is to test a specific hypothesis derived from a psychological theory, the question of external validity of the findings is irrelevant Moreover, some real-life issues cannot be mimicked at the lab due to ethical considerations (e.g., child abuse and neglect, reactions to extreme stressful phenomena etc.) What if we want to generalize the findings beyond the boundaries of lab? E S TA B L I S H I N G T H E E XT E R N A L VA L I D I T Y O F E X P E R I M E N TA L F I N D I N G S Field experiments: to study and manipulate variables in a natural, real-world setting rather than in a controlled laboratory environment. Ex: Does the presence of visible security cameras in a public park influence people's behavior and perceptions of safety? Campuses, malls, parks, any publica places that people are found Which one is better? External or internal validity? NATURAL GROUPS DESIGN An individual differences variable is a characteristic or trait that varies across individuals Consider political orientation (communist, republican, democrat, socialist) A researcher cannot employ ransom assignment But control political orientation by systematically choosing individuals naturally belong to these groups In other words, the ratio of the Ps having a particular orientation is manipulated Differentiate experiments involving IVs whose levels are selected from those involving IVs whose levels are manipulated NATURAL GROUPS DESIGN Natural groups design: Experiments involving independent variables whose levels are selected—like individual differences variables Used when ethical and practical considerations prevent researchers from directly manipulating IV Ex: The effect of a loss of a parent in early age on the emotion expression Effect of childhood trauma on emotional processing The effect of political orientation on the evaluation of a conspiracy theory Effect of divorce on emotional disorders What about causality in natural groups designs?  OTHER THINGS TO CONSIDER How to get subjects for your Keeping good notes study (Recruiting Ps) Record keeping Subject pool Manipulation Check Volunteers Pilot Testing Minimizing Error Standardizing the procedure Experimenter expectancy bias (double blind) Demand characteristics MANIPUL ATION CHECK Manipulation check: assigning a Effect of social exclusion (via cyber ball game) on the mood scale (PANAS) to measure the emotional information emotions of Ps after the manipulation processing We expect that Ps in the exp group has higher scores in negative To elicit social exclusion feeling, emotions (particularly felt excluded, Ps were assigned to two sad, nervous etc.) different conditions: If that so, then social exclusion Exp group 1: plays cyber ball game manipulation woks, meaning that and excluded when Ps were excluded via cyber ball game, they really felt Control group: plays cyber ball game excluded/isolated and included FOOD FOR THOUGHT Explain what internal validity is and why experiments are considered to be high in internal validity as compared to survey research? FOOD FOR THOUGHT Explain why it is important to standardize the procedure of an experiment and several ways to do this