Psychology Research Methods

Triangulation

Triangulation

Cross-checking of information and conclusions in research by the use of

• different methods

• different sources of data

• different researches

• different perspectives(theory)

Data triangulation

When a researcher ollects data from different sources to confirm the findings from multiple perspectives

Method triangulation

Comparing data from the use of different methods that could include qualitative and quantitative in order to make sure it wasnt the choice of the reserach methods that lead to the findings

Research triangulation

Involves the use of serveral observers, interviewers, or researchers, to compare and check data collection and interpretation

theory triangulation

looking at the data using different theoretical approaches

• biological

• cognitive

• sociocultural

Observation

careful examination of a object, process, or behavior for the purpose of collecting data

Limitations of observations

extraneous variables, artificiality

Experiential observation

researcher actually experiences what the observer experiences

• natural setting

• interactions

• diverse data

reflexivity

the research reflects on their personal characteristics and their relationship to the research setting

Lab observation

observing the behavior of subjects that are in a controlled environment

Naturalistic observation

observing the behavior of subjects that are in a natural environment

Overt observation

participants know they are being watched and studied

Covert observation

participants do not know they are being watched and studied

Participant observation

observer actively engages in the activities of the group or community

non-participant observation

observer remains detached from the group being studied

structured observation

observer uses a predefined framework to collect data, focusing on specific behaviors

unstructured observation

observer does not use a predefined framework, more flexible, open-ended approach to exploring behavior

Event sampling

research only makes note when they observe the behavior being studied

Point sampling

researcher makes note of the behavior of each participant then moves on to the next participant

time sampling

researcher makes note of behavior of the sample at a regular time interval (every three minutes for example)

Structured interview

fixed lists of questions in fixed order

Semi-structured

Only certain questions should be asked

Natural Flow

Clarifications with follow-up question

Unstructured Interview

No plan, participant driven

focus group interviews

direct cotact

sensitive topics

Interviewing process

1. Establish rapport

2. Engage person by asking carefully phrased questions

3. Listen, observe behavior

4. Ask follow-up questions

Open-ended question

A question designed to encourage a full, meaningful answer using the participant’s own knowledge and/or feelings

Close-ended question

A question encouraging a short or single-word answer such as a yes/no response

Contrast questions

allows the participant to compare events and experiences

Evaluative questions

asks about the respondent’s feeling about someone or something

Descriptive question

invite the participant to give a general account of something

Structural questions

used to better understanding the meanings of interviews

Content analysis

to analyze the text in a systematic and rigorous way

Thematic analysis

To analyze recurring themes, contents, and categories

True Laboratory experiment

Takes place in a psychology lab, controls extraneous variables, easy to replicate

Field experiment

ordinary, everyday setting, pure behavior, cant control extraneous variables

Natural experiment

The change in the IV occurs naturally, pure behavior, however these are impossible to replicate

Independent Variable (IV)

What the researcher wants to explore, this is manipulated to see if it has an effec on a particular behavior

Dependent Variable (DV)

Outcome or result that is measured

Extraneous variables

Other variables that may affect the results

Quasi experiment

A quasi-experiment in psychology is a research method that lacks random assignment of participants to groups, making it less rigorous than true experiments. It involves manipulating an independent variable to observe its effects on a dependent variable, but without full control over extraneous variables.

Random Sampling:

Used to gain a representative sample, a sampling technique where every member of the population has an equal chance of selection

Independent measures

Using different participants for each condition of the experiment

Repeated Measures

Using the same participants for multiple conditions of an experiment

Matched pairs

pairs of participants are matched in terms of key variables, such as age and IQ.

Research (Alternative) Hypothesis

A researcher’s guess about what will happen

One-tailed Hypothesis

Predicts the direction of the effect of the IV on the DV

Two-tailed Hypothesis

Does not state the direction of the effect of the IV on the DV

Self-selected sample

Made out of volunteers

Opportunity sample

a group that already exists

Stratified sample

drawing random samples from each subpopulation within the target population

Purposive sample

people with a very specific set of traits

Snowball sample

participants recruit other participants from among their friends and acquaintances

Sampling bias

some members of the populatino are more/less likely to be included than others

Ecological fallacy

when we try to draw conclusions about individuals based on data collected at the group level

Demand characteristics

When participants act differently simply because they know that they are in a study

Expectancy effect

 participants believe that they know what the researcher is looking for or what the researcher is trying to do so they are "helping" the researcher. This is a form of compliance. 

Screw you effect

the participant attempts to discern the experimenter's hypotheses, but only in order to destroy the credibility of the study. May also happen if the researcher comes as arrogant or condescending.

Social desirability effect

 the participant answers in a way that makes him/her look good to the researcher. The participants act in this way in order to protect their self-esteem. This is done to avoid embarrassment or judgement. 

Reactivity

when a participant changes his/her behaviour, simply because they know that they are part of a research. Example if there is a problem solving task, the participants may become anxious

Optimism bias

 causes a person to believe that they are at a lower risk of experiencing a negative event compared to other

Participant variability

is a limitation of a study when characteristics of the sample affect the dependent variable. This can be controlled for by selecting a random sample and randomly allocating the participants to the treatment and control groups.

Artificiality

is when the situation created is so unlikely to occur that one has to wonder if there is any validity in the findings.

Situational variables

aspects of the environment that might affect the participant’s behavior, e.g. noise, temperature, lighting conditions, etc. Situational variables should be controlled so they are the same for all participants.

Participant variables

ways in which each participant varies from the other, and how this could affect the results e.g. mood, intelligence, anxiety, nerves, concentration etc.

Researcher bias

Researcher only sees what they are looking for, being affected by their expectations

Confirmation bias

a researcher searches for or interprets information in a way that confirms a pre existing belief or hypothesis

P-hacking

a researcher tries to find patterns in their collected data that can be presented as statistically significant, without first positing a specific hypothesis

Funding bias

Researchers also need to be transparent about the funding of their research. It is important that researchers reflect on why the funders have funded their research and how the funders may use any findings of the research.

Publication Bias

when published research tends to favor only positive results. Negative results - that is, when  the null hypothesis is retained - are not considered very interesting and therefore are often not published.

Casuality

cause and effect relationship can be established

Bidirectional ambiguity

A limitation of many correlational studies. It is not possible to know if x causes y, y causes x, if they interact to cause behaviour, or whether it is just coincidental and no relationship truly exists.

Validity

The extent to which the research tests what it is supposed to test

Internal validity

 Internal validity is high when confounding variables have been controlled and we are quite certain that it was the change in the IV (not something else) that caused the change in the DV.

External validity

Generalizability of findings in the experiment

Population validity

apply the findings to other people

Ecological validity

apply the findings to other situations

Temporal validity

apply the findings to other time periods

Construct validity

How well the experiment measured the concept

Fatigue effects

A type of order effect where a participant decreases in performance in later conditions because they are tired or bored with the activity.

Interference effects

A type of order effect where the first condition may influence the outcome of the second condition. For example, when giving to sets of words to remember, when a participant remembers a word from the first condition when trying to recall words in the second condition.

Practice effect

A type of order effect where a participant improves in performance in later conditions because practice has lead to the development of skill or learning.

Null hypothesis

The null hypothesis states that the IV will have no effect on the DV, or that any change in the DV will be due to chance. 

Standard deviation

SD is part of the inferential statistics which tell us about the results and the relationship between the IV and the DV. We can draw conclusions from this type of data compared to the descriptive statistics.

• Low SD = low variability 

• High SD = high variability

Nominal data

cannot be quantified, are categories (e.g. car brands, zodiac signs)

Ordinal Data

can be ranked from lowest to highest (e.g. Likert scale results) 

Interval data

intervals are assumed to be equal (e.g. SAT scores)

Ration

zero is fixed and meaningful (e.g.number of correct responses, number of words correctly recalled from a list)

Probability value

A small p-value (typically ≤ 0.05) indicates strong evidence against the null hypothesis, so you reject the null hypothesis and accept research hypothesis.

A large p-value (> 0.05) indicates weak evidence against the null hypothesis, so you fail to reject the null hypothesis.