Research Methods
- Created by: MillieAnderson
- Created on: 29-01-15 16:05
View mindmap
- AS Psychology Research Methods
- Definitions
- Qualitative data - Quantities and opinions, not numerical
- Quantitative data - Numerical quantities that can be measured
- Independent variables (IV) - variables which values don't depend on another variable
- Dependent variables (DV) - Variables which values depend on another
- Control - the management of external factors so DV modification is exclusively due to the IV
- Extraneous variables -Undesired variables which influence the IV + DV relationship
- Cause and effect - the relationship between actions so that they are the result of another
- Correlation Analysis - to investigate the relationship between variables and there strength and direction
- Reliability - how consistent the results are. if similar results occur - high reliabilty
- Validity - *** accurate and representative results are
- Aims - The studies purpose, what is being investigated
- Hypothesis - a precise statement predicting he outcome of the study
- Operationalisation - What is being taken as the IV + DV measure
- Research Methods
- Self Report Techniques- useful to find out opinions. Interviews and questioners
- Observational Methods- natural observations in p's natural environment or controlled observation in artificial setting (lab)
- Research Methods
- Laboratory Experiment - carefully controlled environment where all variables are controlled
- High replicability, high control, high precision. Artificial and demand characteristics
- Field Experiment - takes pace in a naturally occurring environment with a manipulated IV.
- Less artificial and less demand characterises on lab experiments. Extraneous variables more likely
- Natural Experiement - observation studies to asses outcomes and impacts of natural intervention
- P's are less conscious so less demand characteristics. likelihood of behaviour happening again is low. Low replicability
- Laboratory Experiment - carefully controlled environment where all variables are controlled
- Ethical Issues
- Informed consent
- Gaining consent from child's' vulnerable person's parent/guardian
- Deception
- With holding information about the purpose of the study
- Deception should be disclosed to participants prio to giving consent and right to withdraw
- With holding information about the purpose of the study
- Protection from harm
- potential threats to physical, psychology and personal beliefs/values
- Consider study form P's perspective, pilot study and right to withdraw
- potential threats to physical, psychology and personal beliefs/values
- Anonymity and confidentiality
- P's to expect all data to be collected to remain anonymous and securely stored
- P's should be warned prior if confidentiality cannot be guaranteed.
- P's to expect all data to be collected to remain anonymous and securely stored
- Right to withdraw
- Right to withdraw at anytime with out reason and request any data to be destroyed
- P's should be told that they have the right to withdraw prior to the study
- Right to withdraw at anytime with out reason and request any data to be destroyed
- Informed consent
- Sampling Techniques
- Random Sampling
- Every person in a given population has an equal chance of being chosen.Not gaurnette random, small sample can be represetitive
- Opportunity sampling
- Non random when researcher selects P's from a given population. Specific and controlled, good for selecting characteristics, not representitve
- Volunteer sampling
- P's select them selves
- Random Sampling
- Validity
- Internal
- Investigator effects ,demand characteristics (interpret purpose change behaviour),social desirability
- External
- Population, temporal and ecological
- Internal
- Experimental Design
- Repeated Measures
- Same group used for both conditions. Less individual difference but practise - order effects
- Counterbalancing - P's undertaking tasks in different orders
- Same group used for both conditions. Less individual difference but practise - order effects
- Independent Groups
- Individual differnces, no order effects.
- Matched Pairs
- Groups for different conditions are matched on characteristics. no order effects, hard to achieve matched pairs
- Repeated Measures
- Graph Measures
- Bar Charts - separate catergories
- Histograms - frequency on continuous scale
- Scattergrams - correlation data
- Hypothesis's
- Directional/ One Tailed -states which will be higher/lower
- Non-Directional/Two Tailed - only states there will be a difference
- Measures of Central Tendency (identify the most typical score)
- Median - putting all scores in order and identifying middle one. may be more than one, not affected by outliers
- Mean - adding all the scores and dividing my no. of scores. takes all date into account, affected by outliers
- Mode - the most commonly occurring score. Not affected by outliers,docent take all data into account. may not be one
- Measures of Dispersion (how the data is spread)
- Range - subtracting lowest score from highest,higher the range the less representative
- Standard Deviation - the average distance of each score from the mean. How representative the mean is. High deviation, less repetitive
- Designing
- Interviews
- Questionairs
- Observations
- Research Method
- Interveiws
- face to face social interaction. Closed interview - fixed set of responses. Open interview - can deviate off topic with few structured questions
- Questionairs
- Written format, qualitative and quantative
- Once developed, simple and cheap, large amounts of data can be gathered with high distribution. If wording is ambiguous P's can interpret questions, anonymous - social desirbility
- Written format, qualitative and quantative
- Observations
- Trained observers watching P's in certain situations
- Training needed, different interpretations, little control but high replicability
- Trained observers watching P's in certain situations
- Case Studies
- Longitudinal exceptional circumstances, in depth investigations of individuals
- Data has high degree of realism which can challenge theories. Lengthy in depth investigation with low replicability
- Longitudinal exceptional circumstances, in depth investigations of individuals
- Correlation Analysis
- Investigate relationships between two variables
- Can measure many variables and their direction and strength. But correlational not causal so cause and effect cannot be established
- Investigate relationships between two variables
- Meta Analysis
- Combining and analysing the results of many different individul studies
- High reliability
- Combining and analysing the results of many different individul studies
- Interveiws
- Definitions
Comments
No comments have yet been made