research methods

• Range


  The difference between the highest value and the lowest value
• Range


  • Overall spread of all data, easy to calculate, may not be representative if there are extremes at either end
• Standard Deviation


  The spread of data around the mean and the relationship between scores
• Standard Deviation


  • Representative, small SD's mean little variation in the scores, harder to calculate than the range
• Mean


  Likely to be the most representative because all scores are used, harder to calculate than other data, can't be used on nominal data, can be affected by extreme scores
• Mode


  Very easy to calculate, unaffected by extreme values, doesn't use all scores, can be multi-modal which is useless
• Median


  Easier to calculate than the mean, unaffected by extreme values, less representative than the mean because it doesn't use all of the scores
• Nominal data


  Data that is in categories for frequency data, people choosing their favourite foods
• Nominal data


  • Most simple data, measures of central tendency are not appropriate, only percentages or proportions
• Interval data


  Continuous data along a set scale with exact and equal intervals, time, cm
• Interval data


  • The best type of data for analysis, all calculations are suitable
• Ordinal data


  Continuous data along a scale, ranked responses favourite to least favourite
• Ordinal data


  • Mean, median, mode standard deviation and range
• Empirical Method


  Gaining data and information through direct observations, testing or experiments, rather than just thoughts and ideas
• Empirical Method


  • Falsifiability: theories should generate testable predictions that can be proved wrong, the process and data must be objective so that it can be tested
• Paradigms


  A paradigm is a shared set of assumptions about the content and methods of a discipline, a paradigm shift occurs when the dominant paradigm is replaced
• Paradigm shift


  • Stimulus-response learning was replaced with social learning theory
• Objectivity


  When data is collected and interpreted free from any bias through using empirical methods
• Reputability


  To test the reliability of findings, research must be replicated exactly to ensure that the results will be the same or there could be a flawed process
• Hypothesis Testing


  Theories seek to explain observations and enable predictions to be made (hypothesis), once it has been accepted or rejected, the theory can be refined with the results
• Qualitative data


  Expressed in words, thoughts, opinions, subjective, can be coded into numerical data
• Qualitative data


  • Rich, detailed data, higher external validity, difficult to analyse or compare, can be subject to bias
• Quantitative data


  Expressed in numbers, objective numerical data, can be easily analysed as it is with statistical tests
• Quantitative data


  • Comparisons and inferences can be made, less subject to bias, may not reflect real life, ignores important qualitative aspects
• Primary data


  Information obtained first hand by the researcher for the purpose of their experiment, study
• Primary data


  • Specific to research, control over data collection, time consuming to collect, design and analyse the relevant data
• Secondary data


  Data that already exists but is beneficial to the current study hypothesis
• Secondary data


  • Inexpensive, quick, requires minimal effort for the researcher, differing methodologies, open to bias, may not be completely relevant or current
• Naturalistic observations


  • Take place in the real or controlled world where the behaviour occurs
  • Increases reliability with control
• Participant observations


  • The researcher becomes part of the group
  • Reduces demand characteristics
  • Difficult to record all data at once
• Non-participant observations


  • The researcher observes from outside the group
  • Reduces demand characteristics
• Covert observations


  • Participants don't know they are being observed
  • Reduces demand characteristics
  • Difficult to gain consent
• Overt observations


  • Participants know they are being observed
• Behavioural categories


  Operationalised behaviours that observers are looking for and make tally charts for
• Time sampling


  • Behaviour is recorded at pre-set time intervals
  • May miss important behaviours between intervals
• Event sampling


  • Behaviour is recorded every time it occurs during the whole observation
  • Can be difficult when lots is going on at once
• Reliability


  • A pilot observation should be conducted beforehand, observers should be trained in advance so they can correctly identify behaviours
  • Inter-observer reliability can be checked if the two observers have a correlation coefficient of at least 0.8
• Informed consent


  A balance between deception and full consent
• Presumptive consent


  Asking a group who would represent those actually in the experiment
• Retrospective consent


  Gained after the experiment