SE - Research Methods

scientific knowledge

derived through systematic methods, relies on empirical evidence and reproducibility

scientific method

set of assumptions, rules and procedures that scientists use to conduct research, results in the accumulation of scientific knowledge, must be objective and replicable

types of research

basic research, applied research

basic research

answers fundamental questions about behaviour

applied research

investigates issues that have implications for everyday life and provide solutions to everyday problems

research design

scientific method a researcher uses to collect, analyze, and interpret data

descriptive research design

answers questions about the current situation, for example: surveys, interviews

limitations of descriptive research

limited to giving a static picture (cannot answer how a certain behaviour develops, why)

correlational research design

the measurement of relevant variables and an assessment of the relationship between or among those variables, the goal is to uncover variables that show systematic relationships with each other. Spearman and Pearson, regression

Limitations of correlational research design

cannot identify causal relationships, can be explained by another variable

Experimental research design

manipulation of a variable, identifies causal relationships

Limitations of experimental research design

cannot experimentally manipulate all variables

key characteristics of science

systematic, critical, controlled, reproducible, empirical

systematic

organized and methodological approach, uses structured methods

critical

requires skepticism and evaluation of claims, findings scrutinized through peer review and replication, is FALSIFIABLE

controlled

minimizes biases and confounding factors

reproductible

results should be replicable by others, consistency - reliability

empirical

based on observable and measurable evidence

theory

frameworks that explain phenomena and predict outcomes, an organized set of principles explaining observed phenomena

empiricism

knowledge based on observation and experimentation, gathers knowledge through sensory experiences and experiments

scientific reasoning

thinking in terms of abstractions or symbols, about many variables or dimensions at the same time, probabilities and proportions

the process of the scientific method:

1.make an observation

2.invent a hypothesis

3.use the hypothesis to make predictions

4.test

theory

framework within which observations are explained and predictions are made

inductive reasoning

from specific to general, involves an element of probability

inductive research

when there is little to no existing literature on a topic, it is common to perform inductive research because there is no theory to test, 3 stages

3 stages of inductive research

1. observation

2. pattern

3. theory

Example: a low cost airlines flight is delayed, another 20 flights from low cost airlines are delayed, low cost airlines always have delays

limitations of inductive reasoning

a conclusion can never be proven but it can be invalidated

deductive reasoning

from general to specific, deriving of a conclusion by reasoning

deductive research

start with a theory (the result of inductive research), formulate a falsifiable hypothesis, collect data, analyse, decide to reject or not the null hypothesis

limitations of deductive research

can only be true if all the premises set in the inductive study are true and the terms are clear

ethical consideration

a set of principles that guide your research designs and practices, work to protect the rights of participants, validity and integrity

The Belmont Report

US government document to protect the human subjects in research

core principles of the Belmont Report

1) respect for persons - individuals must be treated as autonomous agents, 2) beneficence - secure wellbeing, maximize potential benefits and minimize potential risks, 3) justice - the burdens and benefits of research must be distributed fairly

APA ethical principles

1. beneficence and non-maleficience

2. fidelity and responsibility

3. integrity

4. justice

5. respect for people’s rights and dignity

beneficience and nonmaleficience

have to benefit those with whom they work and take care to do no harm

fidelity and responsibility

psychologists accept responsibility for their behavior and manage conflicts of interest

integrity

promoting accuracy, honesty

justice

fairness and justice to all people

respect for people’s rights and dignity

-,-, and confidentiality

ethical issues:

consent, anonymity, withdrawal, harm, debrief, deception

institutional review board

a committee which checks whether the research design is acceptable

to verify a hypothesis

to empirically test and establish its true value

empiricism (verificationism)

knowledge claim is only scientific and meaningful if it is empirically verifiable through direct observation or experiment

hypothesis testing

a formal statistical procedure used by scientists to investigate their predictive ideas about the world by determining the mathematical likelihood of their observations. successful hypothesis testing is associated with type I error.

steps to hypothesis testing:

1. state hypothesis (null vs alternative)

2. collect data

3. analyse

4. compare results

5. present results

alternative hypothesis

predicts a relationship between variables

null hypothesis

no relationship between variables

role of chance

hypothesis testing revolves entirely around determining whether an observed phenomenon happened due to a true systematic effect or merely by random sampling chance

p value

the probability of obtaining your study’s results if the null hypothesis were completely true, alpha = 0-05. If significant, means that there is less than 5% chance the data occurred by random luck, null hypothesis can be rejected

type I error

false positive, reject the null hypothesis when it is actually true

type II error

false negative, fail to reject the null hypothesis when it is actually false

blind spot of p value

p value cannot tell us how big or important an effect is. have to consider sample size and effect size

effect size

a standardized measure of the absolute magnitude (strength) of an effect or relationship, completely independent of effect size. Cohen’s d

clever hans effect

a horse named hans appeared capable of solving complex arithmetic by stamping his hoof. in reality was highly sensitive to the owner’s body language. solution is to implement double blind design

hawthorne effect

industrial workers showed increased productivity when researchers manipulated lighting levels in a factory. In reality productivity improved across all conditions because participants knew they were being observed

placebo effect

expectations and beliefs alone can trigger biological changes. control groups

operational definition

clearly defines variables for measurement

goal of operationalisation

validity, replicability, generalizability

variable

characteristic that can be measured and can assume different values

categorical variables

qualitative

Nominal

a name, label, or category

Ordinal

variable whose values are defined by an order relation between the different categories

binary variables

dichotomous variables, yes or no outcomes

numerical variables

quantitative variables, values are numbers

interval variables

labels and orders, no true zero (eg temp in celsius)

ratio

labels, order, equal interval, true zero (eg weight)

continuous variables

can assume an infinite number of real values within a given interval, no jumps between numbers, can always go smaller

discrete variables

cam assume only a finite number of real values within a given interval (eg number of dogs can be 1, 2…)

IV

manipulate

DV

measure

control variables

held constant throughout the experiment

confounding variables

hides the true effect of another variable in your experiment

latent variables

a variable that cannot be directly measured but you can represent via a proxy

composite variables

a variable that is made by combining multiple variables in an experiment

reliability

consistency of a measure

validity

accuracy of a measure

types of reliability

test retest, parallel forms, internal consistency - split half, cronbach’s alpha, mcdonald’s omega, AND interrater

improve reliability

increase items, standardize environment, train raters

types of validity

face, content, criterion (concurrent, predictive), construct (convergent, discriminant), internal, external (ecological)

face validity

extent to which measurement methods appears “on its face” to measure the construct

content validity

extent to which the measurement covers all necessary aspects of the concept being measured

criterion validity

extent to which a test score corresponds to a concrete, external, real world standard (the criterion), SUBTYPES: concurrent, predictive

concurrent validity

the test correlates strongly with an established benchmark measured at the exact same time

predictive validity

the test score successfully predicts a specific behaviour or outcome in the future

construct validity

how well the operationalized measure adheres to existing theoretical knowledge about the abstract concepts, SUBTYPES: convergent, discriminant

convergent validity

test scores correlate highly with older, already established tests measuring the same or similar traits

discriminant validity

test scores DO NOT correlate with measures of concepts that are theoretically unrelated

internal validity

the degree to which you can confidently state that the IV caused the changes in the DV, without confounding variables interfering

external validity

the degree to which the study’s overall findings can be safely generalized to other populations, settings or time periods, SUBTYPE: ecological validity

ecological validity

a specific type of external validity focusing on whether the experimental results accurately predict real world phenomena outside of a lab environment

qualitative research

the collection, analysis, and interpretation of non-numerical data, researcher is the data tool

five core qualitative approaches

case study, grounded theory, ethnography, phenomenological research, narrative research, HM: action research

case study

a deep intensive investigation of a specific, bounded subject, across multiple sources of information

grounded theory

an inductive approach where researchers collect rich data on a topic and systematically analyze it to generate a brand new theory that is “grounded” directly in the data

ethnography

an immersive approach where the researcher deeply embeds themselves within a specific cultural group or organization for an extended period to understand their shared behaviors, beliefs and language

phenomenological research

an investigation into the “lived experience” of a particular concept or event, goal is to describe and interpret the universal essence of what a group of individuals experienced

narrative research

an examination of stories told by individuals to understand how they sequentially organize and make sense of their specific life events

action research

researchers and participants collaboratively link theory to practice to drive social change

types of qualitative data collection

observations, interviews and focus groups, open ended surveys, secondary textual data

observations

recording field notes of real world interactions