Lecture 1
September 8, 2023 2:30 PM
THINKING CRITICALLY WITH PSYCHOLOGICAL SCIENCE
MODULE 2 - RESEARCH METHODS
Hindsight Bias: The tendency to believe, after learning an outcome, that one would have foreseen it.
Also known as I-knew-it-all-along phenomenon.
It can be dangerous.
Scientific Methods: In Psychology, scientists rely on scientific methods which are based upon:
◊ Empirical Data = info/data you get from direct study, not based on assumption, hindsight bias, etc.
Which consists of:
◊ Theory (a coherent network of explanatory ideas)
◊ Hypothesis (a testable prediction, derived from a theory)
Operational Definitions: Explicitly defining all terms & procedures so that the study could be replicated.
▫ Theory: We can NEVER prove a theory, because to do that you have to prove every single instance, which generally doesn’t happen.
▫ Research: Conduct study/research, if any FLAWS in the findings, re-do the study.
Therefore, there shouldn't be any flaws in the outcomes of the study to test the hypothesis.
▫ The result one study is just a "piece of puzzle" and a part of the theory-building process.
Research Methods:
❖ Survey Method: Information on people's attitude, believes, behaviors is gathered in a questionnaire or interview.
Advantages: Disadvantages:
Fast & Practical Large sample/subjects required
Efficient Question interpretation problems
Sample
Cheap (if done online) Unmatched, target, audience
Anonymous (so people feel quite comfortable, so they are more honest) Dishonesty/lazy replies (no seriousness/boring/frustrated)
Limitation in 'yes or no' question "mini version of
the main interest"
○ So if your result/outcome is true for the 1000 people you surveyed, but, is it true for more 5000 people? Population
• Population = group of interest.
• Sample = the smaller group that you actually study.
○ To generalize from a sample to the population of interest, the sample must be representative of the population (it must be a "mini version"). "main interest"
○ A sample that isn't representative, is called a Biased Sample.
○ How to get a representative sample? = by Random Sampling.
• Random Sampling: is where each person in the population has an equal chance of getting selected for the sample.
❖ Naturalistic Observation: This method is used if you are interested in describing/measuring people's natural behaviour in their natural habitat.
Advantages: Disadvantages:
People are behaving naturally so that you can Researcher doesn't have control over natural situations.
generalise results.
Time consuming & Disturbances
Misinterpreted behaviour
Bad timings of their usual, typical behaviour which may lead to wrong data.
Researcher's presence can influence their habitat.
○ So before conducting this type of observation, you need to have a BEHAVIOURAL CHECKLIST, because there is so much going on in their habitat,
but you need to only be focused on one type of their behaviour.
○ You need to minimize subject/participant reactivity (your influence in their habitat can change their behaviour):
▪ Video record their behaviour: It is ethical as long as it is being done in public in the presence of the subject & not invading their privacy or
habitat.
□ With video recording, you can replay to observe better, but sometimes you can get biased or make assumptions while replaying.
□ To avoid this, have 2 people observe the video recording to eliminate the wrong data/flaws (INTER-OBSERVER RELIABILITY).
▪ Have a participant observer: They are the participant in the research, while also observing the habitat.
□ Be carful, as suspicious participant's behaviour may also influence a stable habitat.
○ Experimenter Bias (Expectancy): Researcher's expectations could bias the observations. It’s a good practice to use 2 observers & determine
inter-observer reliability (look for consistency in their observations) .
▪ Power of Observation!
PILOT STUDY: Before conducting a research/study, the researcher must conduct a pilot study.
□ It is typically a small-scale version of a larger research project.
□ Its purpose is to test the feasibility of the larger project & also to gather preliminary data that can be used to inform the design of the larger study.
Test -1 Page 1
, Lecture 2
September 15, 2023 2:00 PM THINKING CRITICALLY WITH PSYCHOLOGICAL SCIENCE
MODULE 2 - RESEARCH METHODS
Research Methods:
❖ Experimental method: It manipulates (varies) certain variables/factors and controls others (holds other variables/factor constant).
Advantages Disadvantages
You can infer, cause and effect the subject. Not behaving naturally (This method is usually done in a lab, so the subject may be
nervous or anxious).
You can have control over the other variables or Expensive (paying your subjects, lab, assistants, materials).
situations.
○ There are 2 very important variables/factors to conduct an experiment research:
1. INDEPENDENT VARIABLE (IV): Questions based on
□ The variable that the experiment varies/changes/manipulates to see if it actually affects the behaviour. the above variables,
□ In an experiment, the factor that is manipulated; the variable whose effect is being studied. will come with a
□ The situation that the subject is exposed to. scenario question in
□ Cause. which you have to
identify which is IV &
2. DEPENDANT VARIABLES (DV): DV.
□ The behaviour that’s measured to see if the independent variable (IV) had an effect.
□ In an experiment, a factor other than the factor being studied that might influence a study’s results.
□ Effect.
○ IV is the cause & DV is effect of the cause.
○ A variable that YOU DO NOT NEED/OR WANT in your experiment research:
1. CONFOUNDING VARIABLE:
□ It interferes with the result of the study.
□ It changes with the IV and can effect the DV.
□ So you don’t know what variable is causing behavior and can’t draw a conclusion.
○ So in any experiment, its important for you to have:
1. A Control Group:
□ It is used as a standard for comparison.
□ This group gets the "0" level of IV.
□ All other groups are called "experimental groups".
□ All groups must be equal to one another except for the IV.
□ Control groups are necessary to determine whether the IV is actually the one that caused the effect (data) you get.
2. Placebo Control Group:
(used in drug-related studies where people might have strong expectations about drug effects)
□ A special type of control group that is used to control for subjects’ expectations that could affect results.
□ If the control group gets a placebo, then it’s assumed that all groups (the control & experimental group) are equal in
terms of their expectation, and so, any differences in behavior (memory) can be attributed solely to the amount of
alcohol/rug consumed, and not to expectation.
□ Just because of the expectation of the subject, it effects the outcome.
1. Random assignment to groups:
□ This is different from random sampling.
□ By randomly assigning subjects to groups, you're assuming that all individual difference variables are being evenly
distributed among groups so that the groups end up being equal.
❖ Correlational Method: It measures the degree of relationship between 2 variables.
○ Or in other words, It measures the degree to which 2 variables change together.
○ There's no attempt to manipulate or control variables.
○ Rather, naturally occurring variations on both variables are measured to see if they are related.
○ Correlations help us make predictions.
○ Scatterplot graph:
▪ a graph that shows correlational data; each dot represents one person’s 2 corresponding scores.
▪ The closer the dots are to a straight line, the closer or better they are in relation to predict an outcome.
○ Correlation coefficient:
▪ To know exactly how two variables are related we calculate the correlation coefficient (r) by entering all subjects’ 2 scores
in a statistical formula.
R = ± and a number
Questions will be asked
from this topic, but they
□ The number tells you the strength of the relationship.
wouldn't ask to find the
□ It tells us how strong is the predictability .
value of R.
□ That number is ALWAYS between 0 & 1; Never greater or lesser than 1 or 0 because that’s impossible.
□ 1.00 is a prefect correlation.
□ In psychology; 7, 7.5-8.5 is a pretty good & higher correlation.
□ In psychology; 5-6.5 is a moderate correlation.
□ The + or – sign indicates the direction of the relationship.
□ POSITIVE CORRELATION - the values of both variables changes in the SAME direction (as the value of one variable
increases, the value of the other variable also increases).
□ NEGATIVE CORRELATION - the values of both variables change in the OPPOSITE direction (as the value of one
variables increases the value of the other variable decreases).
○ You cannot infer cause-and-effect:
▪ Because you don’t manipulate some variables and control others.
▪ If two variables are correlated it could be that: The slope goes from down to up - Positive correlation. The slope goes from up to down - Negative correlation.
□ A causes B (e.g. low self-esteem causes depression).
□ B causes A (e.g. depression causes low self-esteem).
□ There’s a 3rd underlying variable that causes both A and B (a biological predisposition causes you to have low self -
esteem AND make you depressed).
EXPERIMENTER EXPENTANCY EFFECTS:
• The experimenter has a hypothesis and expects certain results.
• Sometimes he/she can convey that expectation to subjects in a subtle way, such that the subject ends up behaving according
to expectation, thus confirming the hypothesis.
○ Self-fulfilling Prophecy:
• Your expectation can influence your behaviour & other(s) behaviour.
• And what ends up happening is that other's behaviour confirms your expectations.
• So, your expectation is the hypothesis here, and other's behaviour confirms/results outcome of your hypothesis.
○ How to minimize experimenter expectancy effects/experimenter bias:
• Double-blind Study: Both the subject & the experimenter are unaware of the conditions the subject was randomly
assigned to.
• Minimize the amount of time you spend with the subject to avoid socialising & influencing.
• Online study/experiments where the subject only interacts with a computer.
Test -1 Page 2
September 8, 2023 2:30 PM
THINKING CRITICALLY WITH PSYCHOLOGICAL SCIENCE
MODULE 2 - RESEARCH METHODS
Hindsight Bias: The tendency to believe, after learning an outcome, that one would have foreseen it.
Also known as I-knew-it-all-along phenomenon.
It can be dangerous.
Scientific Methods: In Psychology, scientists rely on scientific methods which are based upon:
◊ Empirical Data = info/data you get from direct study, not based on assumption, hindsight bias, etc.
Which consists of:
◊ Theory (a coherent network of explanatory ideas)
◊ Hypothesis (a testable prediction, derived from a theory)
Operational Definitions: Explicitly defining all terms & procedures so that the study could be replicated.
▫ Theory: We can NEVER prove a theory, because to do that you have to prove every single instance, which generally doesn’t happen.
▫ Research: Conduct study/research, if any FLAWS in the findings, re-do the study.
Therefore, there shouldn't be any flaws in the outcomes of the study to test the hypothesis.
▫ The result one study is just a "piece of puzzle" and a part of the theory-building process.
Research Methods:
❖ Survey Method: Information on people's attitude, believes, behaviors is gathered in a questionnaire or interview.
Advantages: Disadvantages:
Fast & Practical Large sample/subjects required
Efficient Question interpretation problems
Sample
Cheap (if done online) Unmatched, target, audience
Anonymous (so people feel quite comfortable, so they are more honest) Dishonesty/lazy replies (no seriousness/boring/frustrated)
Limitation in 'yes or no' question "mini version of
the main interest"
○ So if your result/outcome is true for the 1000 people you surveyed, but, is it true for more 5000 people? Population
• Population = group of interest.
• Sample = the smaller group that you actually study.
○ To generalize from a sample to the population of interest, the sample must be representative of the population (it must be a "mini version"). "main interest"
○ A sample that isn't representative, is called a Biased Sample.
○ How to get a representative sample? = by Random Sampling.
• Random Sampling: is where each person in the population has an equal chance of getting selected for the sample.
❖ Naturalistic Observation: This method is used if you are interested in describing/measuring people's natural behaviour in their natural habitat.
Advantages: Disadvantages:
People are behaving naturally so that you can Researcher doesn't have control over natural situations.
generalise results.
Time consuming & Disturbances
Misinterpreted behaviour
Bad timings of their usual, typical behaviour which may lead to wrong data.
Researcher's presence can influence their habitat.
○ So before conducting this type of observation, you need to have a BEHAVIOURAL CHECKLIST, because there is so much going on in their habitat,
but you need to only be focused on one type of their behaviour.
○ You need to minimize subject/participant reactivity (your influence in their habitat can change their behaviour):
▪ Video record their behaviour: It is ethical as long as it is being done in public in the presence of the subject & not invading their privacy or
habitat.
□ With video recording, you can replay to observe better, but sometimes you can get biased or make assumptions while replaying.
□ To avoid this, have 2 people observe the video recording to eliminate the wrong data/flaws (INTER-OBSERVER RELIABILITY).
▪ Have a participant observer: They are the participant in the research, while also observing the habitat.
□ Be carful, as suspicious participant's behaviour may also influence a stable habitat.
○ Experimenter Bias (Expectancy): Researcher's expectations could bias the observations. It’s a good practice to use 2 observers & determine
inter-observer reliability (look for consistency in their observations) .
▪ Power of Observation!
PILOT STUDY: Before conducting a research/study, the researcher must conduct a pilot study.
□ It is typically a small-scale version of a larger research project.
□ Its purpose is to test the feasibility of the larger project & also to gather preliminary data that can be used to inform the design of the larger study.
Test -1 Page 1
, Lecture 2
September 15, 2023 2:00 PM THINKING CRITICALLY WITH PSYCHOLOGICAL SCIENCE
MODULE 2 - RESEARCH METHODS
Research Methods:
❖ Experimental method: It manipulates (varies) certain variables/factors and controls others (holds other variables/factor constant).
Advantages Disadvantages
You can infer, cause and effect the subject. Not behaving naturally (This method is usually done in a lab, so the subject may be
nervous or anxious).
You can have control over the other variables or Expensive (paying your subjects, lab, assistants, materials).
situations.
○ There are 2 very important variables/factors to conduct an experiment research:
1. INDEPENDENT VARIABLE (IV): Questions based on
□ The variable that the experiment varies/changes/manipulates to see if it actually affects the behaviour. the above variables,
□ In an experiment, the factor that is manipulated; the variable whose effect is being studied. will come with a
□ The situation that the subject is exposed to. scenario question in
□ Cause. which you have to
identify which is IV &
2. DEPENDANT VARIABLES (DV): DV.
□ The behaviour that’s measured to see if the independent variable (IV) had an effect.
□ In an experiment, a factor other than the factor being studied that might influence a study’s results.
□ Effect.
○ IV is the cause & DV is effect of the cause.
○ A variable that YOU DO NOT NEED/OR WANT in your experiment research:
1. CONFOUNDING VARIABLE:
□ It interferes with the result of the study.
□ It changes with the IV and can effect the DV.
□ So you don’t know what variable is causing behavior and can’t draw a conclusion.
○ So in any experiment, its important for you to have:
1. A Control Group:
□ It is used as a standard for comparison.
□ This group gets the "0" level of IV.
□ All other groups are called "experimental groups".
□ All groups must be equal to one another except for the IV.
□ Control groups are necessary to determine whether the IV is actually the one that caused the effect (data) you get.
2. Placebo Control Group:
(used in drug-related studies where people might have strong expectations about drug effects)
□ A special type of control group that is used to control for subjects’ expectations that could affect results.
□ If the control group gets a placebo, then it’s assumed that all groups (the control & experimental group) are equal in
terms of their expectation, and so, any differences in behavior (memory) can be attributed solely to the amount of
alcohol/rug consumed, and not to expectation.
□ Just because of the expectation of the subject, it effects the outcome.
1. Random assignment to groups:
□ This is different from random sampling.
□ By randomly assigning subjects to groups, you're assuming that all individual difference variables are being evenly
distributed among groups so that the groups end up being equal.
❖ Correlational Method: It measures the degree of relationship between 2 variables.
○ Or in other words, It measures the degree to which 2 variables change together.
○ There's no attempt to manipulate or control variables.
○ Rather, naturally occurring variations on both variables are measured to see if they are related.
○ Correlations help us make predictions.
○ Scatterplot graph:
▪ a graph that shows correlational data; each dot represents one person’s 2 corresponding scores.
▪ The closer the dots are to a straight line, the closer or better they are in relation to predict an outcome.
○ Correlation coefficient:
▪ To know exactly how two variables are related we calculate the correlation coefficient (r) by entering all subjects’ 2 scores
in a statistical formula.
R = ± and a number
Questions will be asked
from this topic, but they
□ The number tells you the strength of the relationship.
wouldn't ask to find the
□ It tells us how strong is the predictability .
value of R.
□ That number is ALWAYS between 0 & 1; Never greater or lesser than 1 or 0 because that’s impossible.
□ 1.00 is a prefect correlation.
□ In psychology; 7, 7.5-8.5 is a pretty good & higher correlation.
□ In psychology; 5-6.5 is a moderate correlation.
□ The + or – sign indicates the direction of the relationship.
□ POSITIVE CORRELATION - the values of both variables changes in the SAME direction (as the value of one variable
increases, the value of the other variable also increases).
□ NEGATIVE CORRELATION - the values of both variables change in the OPPOSITE direction (as the value of one
variables increases the value of the other variable decreases).
○ You cannot infer cause-and-effect:
▪ Because you don’t manipulate some variables and control others.
▪ If two variables are correlated it could be that: The slope goes from down to up - Positive correlation. The slope goes from up to down - Negative correlation.
□ A causes B (e.g. low self-esteem causes depression).
□ B causes A (e.g. depression causes low self-esteem).
□ There’s a 3rd underlying variable that causes both A and B (a biological predisposition causes you to have low self -
esteem AND make you depressed).
EXPERIMENTER EXPENTANCY EFFECTS:
• The experimenter has a hypothesis and expects certain results.
• Sometimes he/she can convey that expectation to subjects in a subtle way, such that the subject ends up behaving according
to expectation, thus confirming the hypothesis.
○ Self-fulfilling Prophecy:
• Your expectation can influence your behaviour & other(s) behaviour.
• And what ends up happening is that other's behaviour confirms your expectations.
• So, your expectation is the hypothesis here, and other's behaviour confirms/results outcome of your hypothesis.
○ How to minimize experimenter expectancy effects/experimenter bias:
• Double-blind Study: Both the subject & the experimenter are unaware of the conditions the subject was randomly
assigned to.
• Minimize the amount of time you spend with the subject to avoid socialising & influencing.
• Online study/experiments where the subject only interacts with a computer.
Test -1 Page 2
