AS Research Methods

AS Research Methods

1. Investigation Design

   • Aims and Hypotheses
     • Aim
       • A statement of what the researchers intend to find out in a research study (a general statement of the purpose of an investigation)
     • Hypothesis
       • A precise and testable statement about the assumed relationship between variables
       • Types of Hypotheses
         • Directional (One-tailed)
           • States the direction the results will go in
         • Non-directional (two-tailed)
           • Is less clear of the direction
         • Null (No difference/effect
   • Experimental Design
     • Repeated Measures
       • The same participants are used in both conditions
         • +Can control participant variables +Not time consuming and difficult +Less participants -Repetition -Fatigue effect -Order effects
     • Independent groups
       • Ppts are randomly allocated to different groups which represent different conditions
         • +No fatigue effect +No repetition +Reduced demand characteristics -Can't control participant variables -Needs more participants
     • Matched pairs
       • Pairs of ppts are closely matched and randomly allocated to one of the experimental conditions
         • +No fatigue effect +No repetition +No order effects +Can control participant variables -Time-consuming and difficult -Not possible to control all ppt variables
   • Sampling Techniques
     • Random
       • A sample of ppts produced by using a random technique such that everyone has an equal chance of being selected
         • +Unbiased +Representative sample +High population validity -Time consuming to select
     • Opportunity
       • A sample of ppts produced by selecting people who are avaliable at the time
         • +Less time to locate ppts -Bias due to low population validity
     • Volunteer
       • A sample who volunteer
         • +Access to a variety of ppts +More representative -Bias due to high motivation from ppts
     • Stratified
       • A sample produced by identifying sub-groups according to their frequency in the population. Randomly selected from the sub-groups
         • +More representative due to proportional and random selection -Time consuming
     • Systematic
       • Obtained by selecting every nth person
         • +Unbiased as it uses an objective system -Not truly unbiased as you have to select a number
   • Reliability and Validity
     • Validity
       • Whether the observed effect is genuine
         • Internal
           • If the outcome is a result of variables
         • External
           • The extent to which the results can be generalised to other settings
     • Reliability
       • How consistent are the results
   • Demand Characteristics and Investigator effects
     • Demand Characteristics
       • Participants are unconciously aware of the aim of the experiment and give desirable answers
         • Social Desirability effects
           • Tending to answer questions to present themselves in a better light
     • Investigator effects
       • Investigator has an effect on participant's performance
   • Ethics
     • Protection from harm
     • Confidentialty
     • Informed consent
       • Deception
         • Right to withdraw

2. Methods and Techniques used in Psychological Research

   • Experimental Methods
     • Laboratory Experiment
       • A controlled environment where the IV is manipulated and participants randomly allocated. To investigate causal relationships between the IV and DV under controlled conditions
         • +Well controlled +Extraneous variables minimised creating higher internal validity +Easily replicated demonstrating external validity -Artificial, materials may lack mundane realism -Participants may not behave naturally
           • ETHICS: Protection from harm, potential deception, debrief
     • Field Experiment
       • Natural Environment where the IV is manipulated, ppts randomly allocated. Investigates causal relationships between IV and DV in natural surroundings
         • +Less artificial usually higher mundane realism and higher ecological validity +Ppts aren't usually aware of being studied -Less control of extraneous variables reducing internal validity -More time-consuming and expensive
           • ETHICS: Difficult to debrief if ppts aren't aware of being studied. Issues with manipulation and recording behaviour
     • Natural Experiment
       • IVs are controlled naturally, the experiment does not manipulate them, ppts not randomly allocated to conditions within experiment
         • +Allows research where IV cannot be manipulated for ethical or practical reasons +Enables psychologists to study real problems therefore increasing mundane realism -Cannot demonstrate causal relationships because IV is not directly manipulated
           • ETHICS: If people have volunteered informed consent is required. Aware of harm
     • Quasi Experiment
       • Investigates the relationship between IV and DV, particularly where the IV is a characteristic of the person
         • +Able to compare between different types of people -Ppts may be aware of being studied -May be an artificial task reducing ecological validity
   • Self-Report Techniques
     • Questionnaires
       • Data collected through the use of written questions
       • +Distributed cheaply +Respondents may feel more willing to reveal personal information on a questionnaire than an interview -Limited collection from people who can read and write -Only filled out by people who are willing to fill them in
     • Interviews
       • A research method involving real time interaction with another individual resulting in the collection of data
       • Unstructured
         • General aims and possibly some questions, able to guide answers
           • +More detailed information obtained. Tailored to specific needs -Requires more skill than a structured interview because the interviewer has to develop questions on the spot -More expensive/time-consuming for specialist interviewers
       • Structured
         • Questions are planned in advance
           • +Easily repeated due to standardised questions +Easier to analyse due to more predictable answers -Comparability may be a problem if the same interviewer behaves differently on different occasions -Interviewer bias
     • Questions
       • Open
         • Invites respondents to provide their own answers rather than select one provided
           • +Provides unexpected answers gaining new insight -Less literate respondents may find the questions difficult -Qualitative data is more difficult to analyse
       • Closed
         • A predetermined range of answers from which respondents select one
           • +Limited range of answers, easier to analyse graphs -May be forced to select unrepresentative answers lacks validity -Acquiscence bias
   • Observational Methods
     • Naturalistic
       • Carried out in an everyday setting, investigator doesn't interfere
         • +Realistic picture of natural spontaneous behaviour -Little control of extraneous variables
     • Controlled
       • Behaviour is observed under conditions set by the researcher
         • +Observer can focus on particular aspects of behaviour -An unnatural environment can make ppt's behaviour less natural
     • Covert
       • Observing people without their knowledge
         • +Produces more natural behaviour -Ethics: offer a debrief afterwards
     • Overt
       • Observational studies where ppts are aware of being studied
         • +Allows it to be more ethically sound -Behaviour may not be natural
     • Participant
       • Observer participates in activity
         • +Can provide a special insight to the task which may not be gained from the outside -Observer bias
     • Non-participant
       • Observer doesn't join in with the experiment
         • +More likely to be objective because they are not joining in -Lacks insight from ppt
   • Types of Variable
     • Independent (IV)
       • An event directly manipulated by an experimenter in order to test its effect on another variable
     • Dependent (DV)
       • A measurable outcome of the independent variable in an experiment
     • Extraneous (EV)
       • Variables which may/may not be controlled and may/may not affect DV

3. Data analysis and presentation

   • Quantitative data
     • Histograms
       • True zero, no gaps, same size width of bars, continuous line of data
     • Bar Charts
       • Used to represent the data, categorical data
     • Scattergrams
       • Line of best fit to work out the average, correlation between two variables
   • Standard Deviation
     • Central Tendency
       • Mean
         • Statistical average, calculated by adding the numbers and dividing by the number of numbers
       • Median
         • Middle value
       • Mode
         • Most occurring number
     • The spread of scores around the mean
       • If it is large many of the data points are away from the mean
         • If it is small the data was close around the mean
   • Correlation Analysis
     • Designed to test the strength and direction between two variables
       • Strength is expressed by the correlation coefficient which is always between +1 and -1
         • The + is either a positive or negative correlation
           • The number is the strength, the closer to 1 the stronger, closer to 0 is weaker
       • +Easily repeated +Shows trends and relationships -Fails to consider intervening variables -Can't establish cause and effect