Research methods

• Interval data


  Data we can tell a difference between and there is no absolute zero
• Psychologists always test the null hypothesis
• Independent variable


  Variable manipulated by the researcher in an experiment
• P = 5% (0.05) means the probability of chance affecting the result is 5 in 100 or 1 in 20.
• Nominal data


  Data that can be categorized.
• Mean


  Sum of all values divided by the number of values
• Variables

  1. Independent variable

  2. Dependent variable. 3. Extraneous variable. 4. Confounding variables
• Mode


  Most frequently occurring value in a set of data
• Levels of data

  • Nominal
  • Ordinal
  • Interval
  • Ratio
• Null hypothesis assumes there will be no effect in the population
• Ways of controlling variables

  1. Randomisation

  2. Structured standardised procedure
• Directional hypotheses are used when there is previously published research that suggests what will happen
• Ordinal data


  Data that can be placed in order
• Dependent variable


  Variable measured to obtain results in an experiment
• Level of Significance in psychology is P = 5%
• Median


  Middle value in a set of data when arranged in order
• Non-directional hypotheses are used when there is no previously, when you can’t certain what the results will be.
• 5% is accepted hypothesis because it is suitable for general research and is a reasonable balance between a Type 1 error and a Type 11 error
• Reporting a sign test: The observed value of S was compared to the critical value for a one/two-tailed hypothesis
• Measures of dispersion: Range, Standard deviation
• Two types of inferential test: Non-Parametric, Parametric
• Non-Parametric tests make no assumptions and cover all levels of data E.g. Mann-Whitney U test, Wilcoxon matched pairs, Chi squared, sign test
• Type 11 error: Accepted the null when it should have been rejected
• How to report a sign test:

  ‘the observed value of S was…. The critical value for a (one/two) tailed hypothesis, where P=5% and N=…, was ….. As the observed value (…) was (greater/less than) the critical value (…) the result (was/wasn’t) significant and the null hypothesis was (accepted/rejected).

  OBSERVED > CRITICAL NOT SIGNIFICANT SO NULL ACCEPTED.
• P≤5% means the probability is equal to or less than 5 chances in 100 or 1 chance in 20
• Type 1 error: Reject the null when it should have been accepted
• Parametric tests advantage: statistics are more powerful and robust
• Parametric tests require strict criteria about the data to be fulfilled
• Measures of central tendency: Mean, mode, median
• Probability is the likelihood that an event will happen