Research Methods

• Quantitative data

  Information that can be counted or expressed numerically, can be represented in graphs, histograms, tables, charts, interested in frequency and use of experimental method
• Laboratory experiment

  Standardise procedure/ instructions and particpants are aware of the study, involves the manipulation of IV and measurement of effect it has on DV
• Strengths of Lab experiment

  - Provides clear and concise cause/effect relationships

  - Tight controls provide greater accuracy in measurement of effect IV has on DV

  - Easily replicated due to Standardised procedure
• Weaknesses of Lab experiment

  - Artificial environment, people unlikely to act how they would in a normal environment, lacks mundane realism

  - Control of all extraneous variables is hard and random one will influence results

  - Results likely to be effected by sampling, demand characteristics and observer bias
• Field Experiment

  Attempt to measure cause and effect relationships in a participants normal setting
• Strengths of field experiment

  - Improved ecological validity

  - Reduced demand characteristics
• Weaknesses of field experiment

  - More difficult to establish high levels of control so it is more difficult to remove the effect

  - Very difficult to generalise findings to real life situations (different from experiment)

  - Ethical issues as participants are unaware of the experiment
• Variable

  Something you are measuring and something that changes

  In an experimental design, you are measuring the effect an IV has on the DV while controlling EV's
• Independent variable

  Variable that you control and change

  Want to see if it affects something
• Dependent variable

  Variable that changed due to IV

  The one you measure
• Extraneous variable

  Any variable which may confound the DV
• Operationalised variables

  Process of defining what is going to count and the IV and how you are going to measure the DV

  Needs to be clear on what is tested to count as a result

  Know what is measured and how variables are measured
• Hypothesis

  Testable statements which express what the study is investigating, either rejected or accepted to indicate success

  
  

  It will predict the effect a fully operationalised IV will have on a fully operationalised DV

  
  

  They can be one tailed (predict direction), two tailed (change but unsure of direction), directional or non-directional
• Null hypothesis

  States the results will not show a different and any difference is due to chance

  Try to disprove prediction
• Alternative/experimental hypothesis

  Testable statement that proposed outcome of study

  Proposes there will be a difference in some measurable outcome between 2 conditions which are controlled or observed by a researcher
• Controls

  An experiment must reduce the impact of EV's in order to limit the effect on the DV
• Situational variables

  Variables other than IV which may have effected DV from the research setting eg background noise
• Participant variables

  Participants can have an influence on the DV, eg emotions on the day
• Demand characteristics

  Features of experiment inform participants about the aim of study

  They may act in a way they believe is expected and bias the results

  Results are then to do with participants expectations rather than the influence of IV

  E.g, figure out you are the control group and try less hard
• Research design

  How you introduce participants to the variables, try to hold all variables constant except IV which we manipulate to see the effect on DV
• Independent Measures

  2 different groups

  One experiences experimental condition and other acts as the control group so you can compare the differences
• Strengths of independent measures

  - No order effects

  - Design can be used in any experiment
• Weakness of independent measures

  - Subject variables and personal differences

  To fix: allocate people randomly to the groups, pretest and pair results to each group, match for qualities to have equal groups
• Matched Pairs

  A type of IM design, conduct a trial test before to make groups matched
• Strengths of matched pairs

  - Overcome participant variables

  - Controls participant variables
• Weaknesses of matched pairs

  - Hard to find appropriate way to pretest participants

  - Very time consuming
• Repeated measures design

  One group experiences all condition, both groups composed of some people, compare average score on one with average score of other
• Strengths of repeated measures

  - It eliminated effect of individual differences

  - Fewer subjects are required
• Weaknesses of repeated measures

  - Causes order effects

  - Limited use, can't use if participants in one will effect another

  To fix: counterbalance- balance out order of effects, get participants to carry out the condition in a different order
• Descriptive statistics

  Researcher will attempt to analyse results by looking at patterns in data and determine a scale
• Scales if measurement

  Type of scale against a variable is measure
• Nominal data

  Data that can be described and organised into categories or frequences

  eg yes or no, pass or fail
• Ordinal data

  When info can be described in rank order

  eg scale of recall from 1st to 10th
• Interval/ratio data

  Data can be described using a scale that has equal intervals between units

  eg seconds, minutes, hours
• Measures of averages

  Averages tell us the central tendency of a score

  eg mode, median, mean
• Measures of dispersion

  Dispersion tells us how spread out scores are around the central tendency

  (distance/ratio from highest to lowest scores and deviations around the central tendency)

  eg variation ratio, range, standard deviation
• Mode

  The most frequent

  Strengths:

  - Not influenced by extreme scores

  - Useful in showing most popular

  Weaknesses:

  - Crude- most frequence does not equal average

  - Not useful uf many equal modes
• Median

  Middle value when scores are ranked in order

  Strength:

  - Not affected by freak results

  Weakness:

  - Affected by small sample sizes
• Mean


  Mathematical average score (add all scores / total scores)

  Strength:

  - Most sensitive measure

  Weakness:

  - Affected by freak results
• Variable ration

  - Calculates % of score that are not mode

  - Find the modal score, add total number of scores and calculate how many scores are not the mode

  - Divide that by total number of scores

  - x100