SEARCH
You are in browse mode. You must login to use MEMORY

   Log in to start

level: Data anaylsis

Questions and Answers List

level questions: Data anaylsis

QuestionAnswer
Descriptive statistics meaning (part of results)- Measures of central tendencies
Descriptive statistics (maths involved)- Averages (mean, mode, median) - Charts and graphs - How spread out the scores are - dispersion (range)
Calculating dispersion- Standard deviation = variance squared
Inferential statistics meaning- The probability that the result is significant compared to chance
The 5 tests- Sign test (Binomial) - Chi squared (x2) - Wilcoxon test - Mann-Whitney 'U' test - Spearman's Rho
Types of data- Nominal (tally, category) - Ordinal (rankable, interval - time, weight)
Nominal data- Sign test - Repeated measures + matched pairs - Chi squared - Independent measures
Ordinal data- Wilcoxon - Repeated measures + matched pairs - Mann-Whitney U - Independent measures - Spearman's Rho - Correlation
Formula for Mann-Whitney's U testU = RA - (nA (n + 1)/2)
What is the Table Of Critical Values- Used to see if the U score is significant or not - Whether it agrees with the null hypothesis or not
Type 1 Error- When you reject the null hypothesis when it should have been accepted
Type 2 Error- When you accept the null hypothesis when it should have been rejected
Statement of significance (example)P < 0.05 (two-tailed) P = the probability of the results being due to chance
Quantative data- Information about quantities therefore numbers
Quantative data strengths + weaknesses- Easy statistical analysis of difference - Data does not tell us why there is a difference
Qualitive data- Descriptive - Observed but not measured
Qualitive data strengths + weaknesses- Increases validity - More time-consuming - Difficult to draw comparisons
Demand characteristics- The situation where the results of an experiment are biased - When participants beliefs influence the outcome
How to minimise demand characteristics- The blind technique - using deception - Have more than one experimenter