Research Methods

• Aim - outlines the purpose of the research and makes the study clear
• Hypothesis - a testable statement which predicts what will happen in an experiment. Should be clearly DEFINED and OPERATIONALISED
• Types of Hypotheses:

  Directional Hypothesis - states there will be a different and the direction of effect

  Non - Directional Hypothesis = states that there will be a difference but NOT the direction of the effect

  Null Hypothesis - states that there will be no difference
• Types of Variables :

  • Independant Variable - the variable that is manipulated and assumed to have a direct effect on the DV.
  • Dependent Variable - the variable you measure.
• Levels of the IV

  • When experimental groups are used within experimental conditions to make a comparison
  • Control Condition = no change is made
  • Experimental Condition = In which a change is made to the IV.
• Control of Variables

  • Extraneous Variables - unwanted variables within an experiment that can alter or effect the end result. (e.g. Light, temperature)
  • Should ALWAYS be removed to avoid. Will always be straightforward to control.
• Types of EV.

  1. Participant Variables
  2. Participant Effects
  3. Investigator Effects / Bias
  4. Situational Variables
  5. Order Effects
• Extraneous Variable : Participant Variables

  • Participant Variables - characteristics of participants (e.g. age, gender, age, personality

  Control = use repeated measures design and allocate participants carefully
• Extraneous Variable : Participant Effects

  1. Hawthorne Effect - participants perform better when observed
  2. Screw you Effect - participants deliberately behave in a way to spoil an experiment
  3. Demand Characteristics - participants will act in a certain way they think will be more demanded in an experiment
  4. Social Desirability Bias - participants want to be seen in the best way possible so they behave differently

  CONTROL - single blind technique = the participants don't know the aim of the experiment
• Extraneous Variable: Investigator Effects Bias

  • the investigator will may effect how the participants are behaving because they want to achieve a certain outcome from the experiment. May start to act Bias to achieve this.

  CONTROL : use standardized instructions and Double Blind Technique ( researcher and participants don't know the aim of experiment )
• Extraneous Variable: Situational Variables

  • Situational Variable - a feature of the research setting may act as a cofounding variable (e.g. temperature , time of the day , lighting)
  • CONTROL - use standardized procedures and instructions
• Order Effects - participants may perform differently on two different occasions, due to physical or mental condition ( e.g. tired, bored )

  CONTROL - Counterbalancing (ABBA = 1/2 do 1 part of an experiment and then they swap)
• Cofounding Variables

  • variables that change systematically with the IV. ( e.g. If someones personaility has changed it may effect the experiement but we are unable to control it )
• Pilot Study - when a practice of the actual experiment is done to test for any errors or mistakes that may occur.

  !!! - people that were used in the practice experiment can not be used in the actual experiment because they know the aim of the research.
• Validity and Reliability

  • Validity - the extent to which something measures out what is set out to measure
  • Internal Validity - refers to the extent to which the results of a study occur due to the variable being tested and not any other variables
  • External Validity - the extent to which results can be generalised to different situations and life
• Improving Validity

  • Internal - can be done through tighter control on Extraneous Variables
  • External - can be done through realistic test and natural settings
• Reliability - refers to the research findings are consistent. This means if the experiment results are done again, it should be come out with the same results.
• Improving Reliability

  • Improving objectivity of measures = will increase consistency overtime.
  • Defining Variables = improves reliability
  • Standardise procedures
• Sampling

  • the process in which Ps are selected. There are 5 categories :

  1. Stratified
  2. Opportunity
  3. Volunteer
  4. Random
  5. Systematic
• Random Sampling - every member of a target population has the chance of getting picked

  STRENGTH - reduces risk of researcher bias

  LIMITATION - 1. Time consuming 2. Can still produce a bias result which means you can't generalise
• Systematic Sampling - when the nth person of a population is picked (e.g. every 10th person )

  STRENGTH - researcher has no influence on who is picked

  LIMITATION - Time Consuming
• Stratified Sampling - when key characteristics of the population are being reflected in the sample

  STRENGTH - 1. No researcher Bias 2. Generalisation is more possible

  LIMITATION - 1. Time consuming 2. Complete representation cannot be taken as only the people characteristics have been chosen
• Opportunity Sampling - participants where were available at the time of the study.

  STRENGTH - Quick and Convenient as you can pick anyone

  LIMITATION - Research Bias is most likely to occur as you'll pick people who are most likely to do well in the experiment
• Volunteer Sampling - participants offer to be part of the study and are selected by the researcher.

  STRENGTH - Participants will be willing to take part meaning they will put effort

  LIMITATION - method will produce a bias sample as only participants will only volunteer if they know they will do well
• Experimental Designs - ways in which participants can be organised, in relations to experimental conditions. There are 3 experimental conditions:

  1. Matched Pairs
  2. Repeated Measures
  3. Independent Groups
• Experimental Design : Independent Groups

  • Different Ps in each condition
  • Ps are randomly allocated to experimental groups
  • Each group experiences different levels of IV
  • Performances are compared

  STRENGTH - 1. No investigator bias as Ps were randomly allocated. Allows for high external validity

  LIMITATION - More Ps are required to do this = more time and money
• Experimental Design : Repeated Measures

  • Some Ps are used in each condition
  • Order of conditions need to be counterbalanced to avoid order effects
  • Data is compared to check for any differences

  STRENGTH - 1. Same Ps are used in each condition 2. Fewer Ps are needed = less costly

  LIMITATION - Order effects may occur as participants are doing task more than once
• Experimental Design : Matched Pairs

  • Ps are matched in terms of similarity ( e.g. age, sex ) into pairs
  • One P from each pair go into a certain condition and then switch

  STRENGTH - No order effects as participants are doing once

  LIMITATION - more costly as twice the amount of people are needed AND time consuming