research methods year 2

• what is content analysis?

  a research technique which enables the indirect study of behaviour by examining communication that people produce e.g. in texts or media
• what is coding?

  the stage of content analysis in which the communication to be studied is analysed by identifying each instance of the chosen categories e.g. every time a derogatory words for mental health sufferers is used in a newspaper
• what is thematic analysis?

  an inductive and qualitative approach to analysis which involves identifying implicit or explicit ideas within the data
• what is reliability?

  refers to how consistent a measuring device e.g. if a particular measurement is made twice and produces the same result then the measurement is reliable
• what is validity?

  the extent to which an observed effect is genuine - does it measure what it's supposed to measure and can it be generalised beyond the research setting within which it was founded?
• what is test-retest?

  repeating the studies using the same procedures at different times and then test the correlation between the two versions
• what is inter-observer reliability?

  two or more observers record behaviours during the same observation using the same behavioural categories, then they test the correlation between each tally of behaviour to identify if the behavioural categories are appropriately operationalised
• how can you improve reliability with questionnaires?

  use closed questions to reduce the range of possible responses, and if there's an already established questionnaire then use it
• how can you improve reliability with interviews?

  use structured interviews to ensure each ps has a similar experience, and the researcher can compare the answers
• how can you improve reliability with observations?

  improve the training given to observers can improve their accuracy of assigning behaviour to the correct categories, pilot studies can identify poorly designed behavioural categories
• how can you improve reliability with experiments?

  use standardised procedures for each ps, use established tests as measures rather then creating a new test
• what is face validity?

  whether the test appears to be testing what it claims to be testing
• what is concurrent validity?

  the extent to which data from the newly created test is similar to an established test
• how can you improve validity in experiments?

  use a control group to better assess whether changes in the dv were due to the iv, standardise procedures to minimise the impacts of participant reactivity and investigator effect
• how can you improve validity in questionnaires?

  incorporate a lie scale within the questions to assess the consistency of a ps's response and to control for the effects of social desirability bias
• how can you improve validity in observations?

  using covert observations to avoid demand characteristics, don't use behavioural categories which are too broad or ambiguous
• how can you improve validity in qualitative research?

  researcher needs to present interpretive validity (when the researcher's interpretation of events matches that of their ps)
• what is a null hypothesis?

  suggests there's no difference/relationship between the two conditions
• what is an alternative hypothesis?

  suggests there is a relationship between the two variables being studied
• what is objectivity?

  data should be collected and interpreted in ways that avoid bias, meaning that the data is not influenced by the researcher's opinions or expectations
• what is reliability?

  scientist's are required to carefully record their methods and produce standardised procedures so that others can repeat the experiments and observations
• what is falsifiability?

  a theory needs to be constructed in a way that it can be empirically tested (demonstrates it's not true)
• what are paradigms?

  a set of established assumptions
• what is a paradigm shift?

  scientific fields develop in a series of 'scientific revolutions', when enough evidence is collected which contradicts the current paradigm, the community moves to a new one
• What are statistical tests used for?

  to determine whether a difference or an association/correlation can be found in a particular investigation is statistically significant
• what are the three factors used to decide what statistical test to use?

  -whether the researcher is looking for a difference or a correlation?

  -in the case of a difference, what experimental design is being used

  -the level of measurement
• what is nominal data?

  data is represented in the form of categories
• what is ordinal data?

  data which is ordered
• what is interval data?

  data based on numerical scales that included units of equal, precisely defined size
• what is a sign test?

  a statistical test for a difference in scores between related items. Data should be nominal level or better
• what is spearmans's rho?

  a test for a correlation when data is at least ordinal level
• what is pearson's r?

  a parametric test for a correlation when data is at interval level
• what is wilcoxen?

  a test for a difference between two sets of scores. Data should be at least ordinal level using a related design
• what is Mann Whitney?

  A test for a difference between two sets of scores. Data should be at at least ordinal level using an unrelated design e.g. independent groups
• what is a related t-test?

  a parametric test for a difference between two sets of scores. Data must be at interval level with a related design e.g. repeated measures of matched pairs
• what is an unrelated t-test?

  a parametric test for a difference between two sets of scores. Data must be at interval level with an unrelated design e.g. independent measures
• what is chi squared?

  a test for an association between two variables or conditions. Data should be at nominal level using an unrelated design e.g. independent measures
• what is a null hypothesis?

  A null hypothesis states there's no difference between the conditions
• what is significance?

  a statistical term that tells us how sure we are that a difference or correlation exists. A 'significant' result means that the researcher can reject the null hypothesis
• what is the usual level of significance?

  0.05, 5%, p_<0.05