Unit 1 - AOS3 - Research Methods.... Flashcards
(27 cards)
Controlled experiment
Investigates cause and effect relationship between 2 or more variables, by manipulating one variable and observing impact on other variable(s).
Independent variable
Condition that is manipulated by researcher to study the effect on another variable.
Dependent variable
A variable that is measured to identify the impact of the manipulaton of the independent variable, on the participant’s behaviour.
Hypothesis - Define and formula
Testable prediction about likely outcome of experiment. In this instance it contains:
- the population of interest
- the independent and dependent variables
- a directional prediction about the interaction between those variables
It is hypothesised that the Population who (IV) will/will be (Prediction) on the (Dependent Variable) than those who (IV).
Population - Define
The entire group of individuals with a specific characteristic that is of interest to the researcher
Sample - Define
Group of participants that are select from, and considered representative of, the population.
Control Group
Group that is exposed to control conditions, without presence of IV
Experimental Group
Group that is exposed to the experimental conditions, with the presence of the IV
Extraneous Variable
Any variable other than the IV which influences the DV - it could include participant differences, the experimenter effect, the order effect, etc.
Confounding Variable
An uncontrolled extraneous variable that causes significant changes in the DV. Its’ effect may be confused with the IV.
3 Sampling Methods + Assess benefits and disadvantages of each
Convenience: Based on the researchers’ accessibility to the population and the participants availability. It is quick and inexpensive, and means participants are less likely to withdraw. However, the sample is not very representative of the population.
Random: Every member of the population has an equal likelihood of being chosen. Eg. done through drawing names out of hat. It is more representative, without researcher bias. However, the sample could skew to a particular group in society.
Stratified: Members of the population are divided into specific groups with individuals who share specific characteristics, and then participants are chosen randomly from those groups in the proportion that they appear in the broader population.
It is highly representatitive, however it can be time-consuming and expensive.
Types of experimental designs - Assess benefits and disadvantages of each
Between groups:
- Randomly assigning participants to the experimental or control group (removes researcher bias, time effective, but does not account for participant differences)
- Pairing individuals based on their shared characteristics and randomly assigning a member from each pair to the experimental or control group (minimises participant differences, but requires some form of pretest)
Within groups:
- One group of participants is exposed to both the control and experimental conditions (eliminates participant differences completely, but could result in order effect - addressed via counter-balancing)
Mixed-methods design: - A combination of within groups and between groups, which may test the effect of the independent variable pre and post time period (participant differences are controlled for, however, there is less control over participant knowledge of the study)
Participant expectations and their influence on study
This is the way that participants may unknowingly adjust their behaviour due to a belief that the ‘treatment’ provided will be successful.
It could act as an extraneous variable, where 5 minutes of meditation each morning may not affect their sleep, but their belief that it will does.
Placebo
A fake drug or treatment to induce the belief that participants are in the experimental group - i.e a sugar pill
Placebo effect
A change in a participants’ behaviour due to a belief over the impact of participating in an experiment
Single-blind procedure
Participants are unaware of who is in the experimental or the control group
Double-blinded procedure
Both the participants and the researcher are unaware of who is in the experimental or control group - removes experimenter effect and participant expectations
Methodologies of data collection
Observational study/Fieldwork: Researcher observes and records behaviour of individuals in their natural environment - eliminating artificality, but posing the risk of ‘observer bias’
Correlational study: Examines the relationship between two or more variables, without manipulation. It does not seek to establish a causal link.
Product, process, system development: Something new is designed to meet a social need, allowing practical application of research. However, it can be challenging to apply globally.
Case Study: Investigation of a specific activity, behaviour, event or problem containing real-world complexities. It is typically used only when there is a small/limited sample size - Genie.
Classification and identification: Involves two distinct components - arranging phenomena into different sets, and recognising phenomena to belong into a new set. - helps identify abnormal behaviours.
Literature Review: Collecting and analysing secondary research, identifying patterns, deviations, controversy. It is often used as the starting point for primary data research.
Primary Data - examples
Data a researcher gathers themself, during a current study - i.e through fieldwork
Secondary Data - examples
Previous research or knowledge used by researcher in their own study - i.e literature review
Data modelling distinguished to Simulation
Data modelling involves a static or virtual representation of a situation, whereas simulation is the active component to the model, using it to run a specific scenario.
Subjective Data
Based on matters of opinion and interpretation, that can be difficult to analyse but provides insight into i ndividual beliefs
Objective Data
Data that can be observed and measured, which is easy to compare but does not provide reasons behind the situation
Qualitative vs Quantitative
Qualitative: Changes in the quality of behaviours, usually expressed in words - challenging to analyse statistically
Quantitative: Data collected in systematic, controlled procedures which is typically presented in numerical form - restricts from futher detail