Flashcards
6) Inferential statistics
Cards in this deck (39)
- 1In a normal distribution curve, where should the mean mode and median lie?In the middle ( the same point)
- 2What are the three levels of measurement?-Nominal
-Ordinal
-interval - 3Describe nominal data-Two or more categories
- Shows the number of times something (a behaviour) occurred
-Least meaningful - 4Describe ordinal data- Slightly more meaningful
-Rankings - 5Describe interval data- Much more meaningful
-Known measurements e.g. cm, grams - 6What does a small variance tell us about the spread of scores around the mean?The results are close to the mean
- 7What does a large variance tell us about the spread of scores around the mean?The results are far away from the mean
- 8What does the standard deviation tell you?Tells us how spread out the numbers are from the mean
- 9What is a strength of variance?-Not distorted by outliers or extreme scores
- 10What is a weakness of variance?-May hide some of the characteristics of the data
- 11What is a strength of standard deviation?-Easier to direct judgements
- 12What is a weakness of standard deviation?-This is not useful when the data is not normally distributed
- 13what is probability?How likely something is to happen and is expressed as a number between 1-10
- 14What is a significant result?A significant result is one where there is a low probability that chance factors were responsible for any difference
- 15What is a type 1 error?-where you think that there was an effect but there actually wasn't
-Too lenient
-This would mean that we would reject the null hypothesis and accept the research hypothesis
-When actually we should have accepted the null hypothesis and rejected the research hypothesis - 16What is a type 2 error?-Where we think that there was no effect but there actually was
-This would mean that we would accept the null hypothesis and reject the research hypothesis
-when actually we should have rejected the null hypothesis and accepted the research hypothesis. - 17What do we do if the probability the results are due to chance is less that or equal to 5%?-Reject the null hypothesis
-Accept the research hypothesis
-There is significance in the results - 18What would we do if the probability the results are due to chance was more than or equal to 5%?-Accept the null
-Reject the research hypothesis
-There was no significance - 19What are the three criteria for a PARAMETRIC test?-Populations drawn from should be normally distributed
-Should have at least interval data
-Should have no extreme scores - 20What is the criteria for the Mann Whitney U test?-Independant measures
-Difference
-Ordinal/ interval data - 21How would you calculate this?1)Rank scores from lowest to highest
2) Add up the ranks for each condition
3) Now use the formula depending on which ever ranking is LOWEST
4)Compare the scores to the critical angle - 22What would you do if the score is higher thank the critical value?-Accept the null
-Reject the research hypothesis - 23What would you do if the score was lower than the critical value?-REject null
-Accept research hypothesis - 24What is the criteria for the Wilcoxon Signed Rank TEst?-Difference between variables
-Interval/Ordinal data
-REpeated measures - 25How would you calculate this?1) Find the difference between each score in each condition
2) Rank the difference
3) Count how many negative figures and how many positive figures
4)Add up the ranks of whichever is the least frequent (positive or negative) and this is your observed value - 26What would you do if the observed value is LOWER than the critical value?-Null hypothesis is rejected
-Alternate hypothesis is accepted - 27What would you do if the observed value is HIGHER than the critical value?-Accept the null
-Reject research hypothesis - 28What is the criteria for a Binomial sign test?-Difference between variables
-Repeated measures
-Nominal data - 29How you calculate this?1) Positive and negative signs need to be added (e.g. Yes no= +, No yes = -, Yes yes or no no = ignore)
2) Add up all the +ves and -ves
3) use whichever one is lowest as the observed value - 30What would you do if the observed value is lower than the critical value?-There is a significance so...
-Reject null
-Accept research hypothesis - 31What would you do if the observed value is higher than the critical value?-There is no significance so...
-Accept null
-Reject research hypothesis - 32What is the criteria for the Chi square test?-Difference between variables
-Nominal Data
-INdpendant measures - 33How would you calculate the Chi squared test?1) Add the total amount for each column
2) Calculate the expected frequencies (Raw total x Column total/ overall total)
3) Use formula to calculate Chi squared (observed- expected) squared/ expected - 34What would you do if the observed value was less than the critical value?-No significance so...
-Accept null
-Reject research hypothesis - 35What would you do if the observed value was higher than the critical value?-Signifcance so...
-Accept research hypothesis
-Reject null - 36What is the criteria for Spearman's Rank?-Interval/ ordinal data
-Correlation or relationship - 37How do you calculate this?1) Rank the scores OF EACH VARIABLE SEPARATELY
2). Find the differences between them
3) square the differences
4) Add up all of the differences squared
5) Us the formula... - 38What would you do if the observed value is lower than the critical value?-There is no significance so...
-Accept the null hypothesis
-Reject the alternate hypothesis - 39What is different about testing for significance in the Chi squared test and the Spearman's Rank?-If the observed value is lower than the critical value then there is no significance and if the observed value is higher than the critical value then there is significance
-Whereas in other non-parametric tests if the observed value is lower than the critical value there is significance and if the observed value is higher than the critical value then there is no significance
Information
- Description
- AS level AS Psychology (Research Methods) Flashcards on 6) Inferential statistics, created by Roberta Power on 07/12/2016.
- Created by
- Roberta Power
- Views
- 0
- Rating