Psychology Unit 0 (Myers Psychology 4th Edition)

Ethics, Correlation, Statistics, Experiments, Research Terms+Methods

Informed Consent

Participants must be fully informed about the experiment and it’s potential risks/benefit before they agree

E.g: a drug study must inform participants about the risks of the drug

Informed Assent

This is for minors or individuals with limited abilities: consent must be obtained from the legal guardian

E.g: a study working with kids in a classroom must get consent from the guardians

Confidentiality

Information about the participant must be kept private and secure

E.g: a participants information cannot be revealed to the public without redactions

Anonymity

Data should be collected without identifying information

E.g: Participants reaction to a drug can be collected without collecting their demographics, address etc

Deception

May be used to create a realistic setting or avoid biases

E.g: Researchers tell parents they are studying their child, but are in fact studying the parents

Debriefing

Participants must be fully informed of everything that happened, including deception, after the experiment and can ask questions

E.g: researchers debrief the participants on the effects of a drug after experimentation

Correlation

Examines the relationship between two variables

E.g: the relationship between sleep and happiness

Positive Correlation

As one variable increases, so does the other (and vice versa)

E.g: increased sleep, increased academic performance

Negative Correlation

As one variable increases, the other decreases; they are always going in opposite “directions”

E.g: Sleep decreases, irritability increases

Scatterplot

Data plotted on a graph that does not follow an equation; a line of best fit can be drawn

E.g: A scatterplot plots the hours of sleep vs academic performance

Correlation Coefficient

Numerical value that shows the strength and direction of the correlation, on a scale of -1 to 1, where the closer it is to -1/1, the stronger, and the sign shows the direction. 1/-1 is perfectly correlated and 0 means no correlation.

E.g: Sleep and irritability have a correlation coefficient of -0.7. They are strongly and negatively correlated.

Illusory Correlation

Seeing a connection between two things when there is none.

E.g: Taller people are smarter

Regression towards the Mean

Extreme or unusual scores tend to fall back towards the average.

E.g: A student has a mean average of a B for tests in a class. After scoring a high A in a test, their next score falls closer to their mean (like a B+).

Descriptive Statistics

Numerical data used to describe characteristics of a set of data through median, mode, mean, standard deviation

E.g: analyzing a set of data to see if the mean is an accurate representation of it

Measure of Central Tendency

Mean, median, mode. Ways to find a number representative of that data

Mode

Most frequently occurring score

E.g: 1 is the mode for “2,1,1,5,1” because it occurs the most frequently

Mean

Average of the set. Sum of all scores / number of scores. Sensitive to outliers (scores far outside the general scope of the other scores)

E.g: 2 is the mean for “2,1,1,5,1” because 2+1+1+5+1 / 5 = 2

Median

Middle score when data is arranged is order. Less affected by outliers.

E.g: 1 is the median for “2,1,1,5,1” because when arranged correctly as “1,1,1,2,5” 1 is the middle number

Range

The difference between the highest and lowest scores

E.g: 4 is range for the data set “2,1,1,5,1” because the 5-1=4

Skewed Distribution

When the mean is higher/lower than the median and mode. Positive/right: when the mean is higher than the other two, so it is to their right. Negative/left: when the mean is lesser than the other two, so it is to their left. This is caused by outliers.

E.g: In the example data set “2,1,1,5,1” the median and mode are 1, but the mean is 2. There is a positive skew because mean is greater than median and mode. This is caused by the outlier 5.

Measures of Variation

Range and standard deviation

Standard Deviation

The average distance of each score from the mean. Tells us how much scores deviate from the average. Larger: data more spread out, smaller: data less spread out (all scores are closer to the mean value).

Never have to find it

Percentile Rank

The percentage of scores lower than a given score depending on standard deviation

E.g: In a normal curve, if the given score is one standard deviation greater than the mean, the percentile rank is 83.6

Normal Curve

When mean, median and mode are all equal, so the curve is bell shaped.

Inferential Statistics

Numerical data that is analyzed and can be generalized by inferring from the sample the percentage this is probable of the population, using the p-value, common threshold and interpretation

E.g: The findings of a study have an x% probability of being generalized to the whole population

Statistically Significant

Statistical statement about the likeliness of a result occurring by chance

Qualitative Data

Focuses of rich detailed data through interviews, focus groups, observations and case studies that can generate hypotheses and theories regarding social phenomena. However, it has subjective interpretation, is difficult to generalize and resource+time intensive.

Quantitative Data

Objective, precise measurements such as Likert scales (1-5 scales), tests, surveys, physiological measures that can test relationships and generalize findings. However, it can oversimplify phenomena, lack context/depth and is limited.

Experiment

Must have manipulation of variables (independent), measures the dependent variable and has a control and experimental group.

Experimental Group

Group which experiences the experiment, where the independent variable is manipulated.

E.g: The experimental group of a study researching a new therapy program is given the new therapy.

Control Group

Group which does not experiences the experiment, where the independent variable is kept at “normal” circumstances.

E.g: The control group of a study researching a new therapy program is given standard, “normal” therapy, not the new one.

Random Assignment

Participants are randomly assigned to experimental/control group to make sure each group is balanced and similar. This ensures that only the manipulation effects data, not the differences between the two groups.

Independent Variable

The variable that is being manipulated.

E.g: playing music in a room where participants are memorizing words

Dependent Variable

The variable that changes as a result of the manipulation of the independent variable.

E.g: the number of words memorized by the people listening to music vs those not

Confounding Variable

A not manipulated variable other than the independent variable that may affect the data/outcome

E.g: the words participants had to memorize were different for the group listening to music and the one not

Third Variable

This is a potential third variable that can affect the perceived relationship between two correlated variables

E.g: Academic success and happiness are correlated. Third variable can be home environment, which directly affects the first two.

Placebo Effect

This is when a person’s belief that something will help them affects what happens to them rather than the thing.

E.g: Participants are given a new medicine and know they are in the experimental group. Their expectations for that medicine impacts how it affects them.

Longitudinal Design

Examines changes in induvial over time. It is expensive, take long and the participant can leave. It also does not examine cohort effects (cohort: generations or age groups).

E.g: the reading development of a few individuals is studied for seven years

Cross Sectional Design

Examines changes between participants of different ages at the same point in time, examining age related changes. Cannot examine changes or time and cohort effects.

E.g: the reading ability of a few children (all of different ages) is studied at the same time and compared

Cross Sequential Design

A combination of Longitudinal and Cross Sequential. Examines changes to individuals over time as well as participants of the different ages (at the same point in time). Examines cohort effects. May be expensive.

E.g: the reading abilities of a few children of different ages is studied for two years, so both children of different ages and same ages (at one point in time) as well as growth (over time) is examined.

Single Blind

Participants do not know whether they are in the control/experimental group. This way, researchers only see the effect of the manipulation.

E.g: in a new medicine study, participants do not know whether they have received the medicine or placebo pill

Double Blind

Both participants and researchers do not know who is the control/experimental groups. Reduces experimenter effect, researchers only see the effect of the manipulation.

E.g: in a new medicine study, participants and researchers do not know whether they have received the medicine or placebo pill

Research Confederates

“Participants” in an experiment who are actually part of the research team and part of the manipulation occuring.

E.g: researchers observe parents and children in a study, some parents are research confederates

Critical Thinking

Thinking that does accept arguments and conclusions but challenges them by examining biases, assumptions, sources, evidence and conclusions.

Cognitive Bias

A bias in thinking such as shortcuts to make thinking easier or rash assumptions, incorrect conclusions or irrational decisions. This creates assumptions and stereotypes.

E.g: assuming there is a correlation between academic success and happiness without examining third variables

Hindsight Bias

After something happens, an individual feels that they could have predicted it or say they felt differently about it at that point, even if they didn’t. Reality is uncertain.

E.g: your sports team wins and you insist you knew it would

Confirmation Bias

Searching for information to support your views. Not looking at total information, lack of objectivity.

E.g: only watching certain news channel

Overconfidence Bias

Overestimate accuracy of our knowledge and judgements. Your abilities and decisions seem better. Poor decision making skills and arrogance or sense of always being right.

E.g: feeling very confident about a test only to find out you did badly

Theory

An explanation using basic principles and ideas to make an attempt to organize behaviors and predict future patterns.

E.g: theorizing how a population will react to a new program based on their previous tendencies and experiments testing similar ideas

Hypothesis

A testable prediction often preceded or implied by a theory

Falsifiability

The ability of an idea, hypothesis or theory to be disproven through observation

Replicate

An experiment must be replicable/easy to replicate in order to be easy to validate

Operational definition

A precise and objective definition of the variables being measured in an experiment

E.g: the operational definition of intelligence is a certain test is a score on an intelligence test

Case study

Investigation of a single study or group, typically of very abnormal circumstances. Gives detailed insight into a unique scenario but cannot be generalized

E.g: a case study of a schizophrenic patient

Naturalistic observation

Observing subjects in a natural environment without manipulation. Gives data in natural settings but lacks a controlled variable and can be influenced by observer bias.

E.g: observing children in a classroom

Survey

Questionaries or interviews to collect data. Gets data from a large number of people that studies covert behaviors but depends on people’s bias/lies in a survey and the sample it is given to.

E.g: surveying students about their study habits

Population

The large group of people that need to be studied.

E.g: the entire student body of a certain school

Sample

The smaller group of people that will be in the experiment.

E.g: a sample of a 1000 people from 4000 students

Random Sample

A sample where participants are randomly selected.

E.g: randomly selecting student participants by their ID numbers.

Representative Sample

A sample that is representative of the population and takes all different people from it into consideration.

E.g: a representative sample from a school consists of people across all grades, athletes, non-athletes, students with a job and without etc

Sampling Bias

A flawed sampling process that creates an unrepresentative sample.

E.g: Surveying students after school means you only get a sample of the students who stay after school

Convenience Sample

A flawed sampling process that does what’s easiest but results in an unrepresentative sample.

E.g: surveying only other students in your class means your sample only consists of students taking that particular class

Social Desirability Bias

Respondents give answers that are socially acceptable instead of truthful.

E.g: Researchers saying they knew an experiment was unethical even if they don’t believe that

Self Report Bias

A person has flawed memories that create inaccurate information about themselves, so they give out inaccurate, non-objective information about themselves

E.g: a person retelling an argument that occurred has self-report bias because they believe they did the right thing even if they objectively didn’t

Experimenter Bias

The researcher’s expectations influences the outcome of the study

E.g: a researchers knows the medicine being tested in an experiment had specific a side effect on many people, so they began asking others if it had that side effect on them (even though the participants didn’t give them that information initially)