- Created by: Hannah Calvert
- Created on: 01-06-13 20:33
Introduction Inferential tests
Significance and Probability
- Inferential tests provide a means of assessing whether any pattern in data collected is meaningful or significant.
- They enable us to make inferences form the research sample to the population.
- Probability= likelihood that a pattern of results could arise by chance.
- Probability levels represent acceptable level of risk or of making a Type 1 error.
- More important research requires more strigent significance levels
- Type 1 error= null hypothesis rejected when true.
- Type 2 error= null hypothesis accepted when false.
- Significance of observed value determined in table of critical values.
- Degree of Freedom normally the amount of participants in the study.
- One tailed test is a directional hypothesis
- Two tailed test is a non-directional hypothesis
- Significance level is usually p(equal to or more than) 0.05 (5%)
- Different research designs and levels of measurement (nominal, ordinal, interval) require different tests
1 of 10
Inferential tests- Spearman's Rho
- Hypothesis states correlational between two variables.
- Each person is measured on both variables.
- Data is at least ordinal (i.e not norminal)
- With repeated measurres and matched pairs
- Scatter line graph is used
- Observed value is greater than or same as the critical value.
2 of 10
Inferential tests- Chi-Square
- Hypothesis states differences between two conditions or association between two variables
- Data is independant.
- Data in frequencies. (nominal)
- Expected frequencies in each cell must not fall below 5.
- Bar graph is used
- Observed value equal to or greater than the critical value
3 of 10
Inferential tests- Mann-Whitney U-Test
- Hypothesis states difference between two sets of data.
- Independant groups design
- Data at least ordinal (i.e not nominal)
- Bar chart is used
- Observed value is equal to or less than the critical
4 of 10
Inferential tests- Wilcoxon T Test
- Hypothesis states difference between two sets of data
- Related design (repeated measures or matched pairs)
- Data at least ordinal (i.e not norminal)
- Line of scatter graph is used.
- Observed value is equal to or less than the Critical value.
5 of 10
Descriptive Statistics- Central Tendency
- Indicates typical or 'average' score
- Mean = sum of all scores divided by number of scores however it is unrepresentative of extreme scores
- Median = middle value in ordered list of scores. Not affected by extreme scores but not as sensitive of all scores than the mean.
- Mode = most common value. Not useful is there are many modes in a set of scores.
6 of 10
Descriptive Statistics- Measures of Dispersion
- Indicate spread of scores
- Range= difference between highest and lowest score. Not representative if extreme scores.
- Standard Deviation= spread of data around mean. Precise measure but influence of extreme scores not taken into account.
7 of 10
Descriptive Statistics- Graphs
- Bar Chart= Illustration of frequency, height of bar represents frequency.
- Scattergram= illustration of correlation, suitable for correlational data. Indicated strength of correlation and direction (positive or negative).
8 of 10
- Quantitive methods not relevant to 'real life'
- Qualitive methods represent world as seen by individual (subjective)
- Data sets tend to be large but few participants
- Qualitive data connot be reduced to numbers
- Reflexivity indicates attitudes and biases of researcher.
- Validity demonstrated by triangulation.
- Summarised by identifying themes in data.
- Inductive (bottom-up) approach so themes emerge, although sometimes deductive (top-down)
- Iterative process- imposing order on the data, reflexting participants perspective.
- 1) Consider data
- 2) Break into meaningful units
- 3) Code each unit
- 4) Create categories themes
- 5) Check themes using new data set
9 of 10
Qualitive V Quantitive
- Quantitive easy to analyse and produces neat conclusions
- But oversimplifies reality and human experience.
- Qualitive sata represents true complexities of behaviour through rich detail of thoughts, feelings etc.
- But more difficult to detect patterns and subject to bias of subjectivity.
10 of 10