Research methods

Research methods

1. Types of data

   • Nominal
     • Named categories are established by the researcher and an aim is counted when it falls into this category
   • Ordinal
     • This is when data is ranked so that it is possible to see the order of scores in relation to one another
   • Interval/Ratio
     • Precise intervals between scores

2. Inferential stats tets

   • Probability - a numerical measure of chance. It represents how likely it is that something will happen
     • number of particular outcomes/number of possible outcomes = between 0-1
     • Statistical significance = the probability that our findings accurately reflect what is happening in the population outside our data.
   • Hypothesis and significance
     • we run a stats test to calculate the probability that the results we have found do not exist in the population, i.e. they are just due to chance
     • The probability we work to is called the significance level and is normally 0.05. If we are significant to 0.05 we accept our experimental hypothesis. If we are not significant to 0.05 we accept our null hypothesis and reject our experimental hypothesis
       • Type 1 errors - when we accept the experimental hypothesis, when actually it was not reflective of the population. The 5% chance it was not reflective came true. There is a 1/20 chance this will happen
       • Type 2 errors - when the null hypothesis is accepted, and think that there is no relationship in our results that reflect the population. However this is not true and there is a relationship
     • Research (alternative or experimental) VS null
   • What ones do you use?
     • What determines the inferential stats test you use?
       • Whether you are looking for a difference or relationshop
       • experimental or repeated measures design
       • What needs to be considered?
         • 1. number of participants 2. one tail or two tailed hypothesis
     • Observed values
       • Spearmans rho = must be equal or greater than the critical value to be significant
       • Chi squared = must be equal or greater than the critical value to be significant
       • Mann-Whitney = must be equal or less than the critical value to be significant
       • Wilcoxon = must be equal or less than the critical value to be significant
   • What type of data do you have?
     • nominal
       • Chi squared
     • ordinal/interval/ratio
       • Looking for a difference or relationship
         • Relationship
           • Spearman's rho
         • Difference
           • independent or repeated
             • repeated
               • Wilcoxon T
             • independent
               • Mann Whitney U
   • Why are they important?
     • They tell us what is significant and what is fluke (not significant)

3. Coding

   • organising data into categories to look at
     • Top down
       • Thematic
         • 1. have a set of categories you are looking out for 2. search through all your data and select all the quotes that fit your categories 3. pick quotes that represent or summarises many of the quotes and summaries and draw overall opinions on the category
     • Bottom up
       • Grounded
         • 1. read all your data 2. cluster common responses together 3. decide on common themes/categories 4. evaluate quotes and responses in each category and summaries each category or use a quote to represent this
     • Issues
       • reflexivity
         • allowing and embracing subjectivity of the researcher to make data more ecologically valid
       • Triangulation
         • using other R.M's or data to compare with your findings to improve reliability