Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
62 Cards in this Set
- Front
- Back
|
What are non-parametric tests? |
Make few or no assumptions about the shapes underlying he population distribution - distribution free tests Stats tests that aren't based on notion of normative curve Stats tests that don't require homogeneity of variance Used on low rank data as stats based on ranks of observations rather than actual values
|
|
|
List all the non-parametric tests and their parametric equivalents (one does not have an equivalent) |
Chi-square - n/a
Wilcoxon T-test - Within Subjects T-test
Mann-Whitney U-test - Between Subjects T-test
Spearman's Rho - Pearson's Correlation Coefficient
|
|
|
What are the advantages and disadvantages of non-parametric tests? |
A: analyses can be simplistic and easier to complete compared to parametric equivalents Useful for nominal or ordinal data sets Can be used with very small sample sizes Useful if data violates the assumption of normality Useful if data is severely skewed or has outliers Makes fewer assumptions so tests can be more robust than parametric tests
D: less powerful as parametric tests use more of the info available in a set of numbers whereas non-parametric tests typically make use of ordinal info only - as such a large sample size may be needed to find significance. Analyses can be complicated if the sample size is large |
|
|
How do you assign ranks for non-parametric tests? |
Order values smallest to largest. The smallest is rank number 1 and it goes up from there. Two numbers of the same score are assigned a tied rank by calculating the average of ranks. |
|
|
What is the Wilcoxon t-test and what type of data and scale of measurement is used in this test? |
The Wilcoxon T-test gives us info about the direction and magnitude of the difference between pairs of scores. Non-parametric data Data must be at ordinal scale of measurement of higher |
|
|
How do you carry out a Wilcoxon T-test? |
Calculate difference scores, then assign ranks to those difference scores (ignoring plus and minus signs and omitting any with score of zero). Remaining calculations are completed on the ranked data. Plus and minus signs added back in. Add up those with plus signs and those with minus signs. Smaller of the two = T value. If T value is equal to or less than critical value then the result is significant. |
|
|
What's the most appropriate way to describe nominal data? |
A mode |
|
|
What's the most appropriate way to describe ordinal data? |
A median and interquartile range |
|
|
What's the most appropriate way to describe interval level data? |
Mean, standard deviation and standard error |
|
|
What level of data would you analyse with a Chi-square? |
Nominal |
|
|
What level of data would you analyse with a Chi-square? |
Nominal |
|
|
What level of data would you analyse with a Wilcoxon and Mann-Whitney U test? |
Ordinal and higher |
|
|
What level of data would you analyse with a Chi-square? |
Nominal |
|
|
What level of data would you analyse with a Wilcoxon and Mann-Whitney U test? |
Ordinal and higher |
|
|
What level of data would you analyse with Spearman's Rho? |
1 variable at ordinal, the other at ordinal or higher |
|
|
What level of data would you analyse with a Chi-square? |
Nominal |
|
|
What level of data would you analyse with a Wilcoxon and Mann-Whitney U test? |
Ordinal and higher |
|
|
What level of data would you analyse with Spearman's Rho? |
1 variable at ordinal, the other at ordinal or higher |
|
|
What level of data would you analyse with Pearson's Correlation Coefficient? |
Interval and ratio |
|
|
What is Deep Brain Stimulation (DBS)? What does it treat? |
Stimulation of particular circuits in the brain which have been damaged leading to disorders such as essential tremor, Parkinson's, dystonia. It has also been applied to other dysfunction such as depression and other mood disorders. |
|
|
How does DBS work? |
Electrodes implanted into brain whilst patient is awake. Electrodes connected to a pulse generator that is implanted below the skin. Can change settings of frequency, pulse width, amplitude to maximise clinical improvement. |
|
|
How does DBS work? |
Electrodes implanted into brain whilst patient is awake. Electrodes connected to a pulse generator that is implanted below the skin. Can change settings of frequency, pulse width, amplitude to maximise clinical improvement. |
|
|
What are the problems of deep brain stimulation? |
With time and disease progression many symptoms in Parkinson's disease develop even in those with DBS surgery. Doesn't work for all patients. Alleviates some symptoms but isn't the cure. Risk of infection due to major surgery. Possible side effects of mood disturbances. |
|
|
What does the Mann-Whitney test show? |
If two groups perform differently on a task or not |
|
|
What does the Mann-Whitney test show? |
If two groups perform differently on a task or not |
|
|
In the Mann-Whitney U test what does N1, N2 and R stand for? |
N1 - no. of participants with largest sum of ranks N2 - no. of participants with smallest sum of ranks R - largest sum of ranks |
|
|
In words, how do you carry out a Mann-Whitney U test? |
Calculate medians for each group. Rank scores, combining the group's, zero values do count. Total the ranks for each group separately. Do the equation. Table gives two ranges of scores - if your U value lies within one of these two ranges your result is significant. |
|
|
What does calculating correlation show? |
Indicates an association between two variables. More specifically, the strength and direction of a linear relationship between the variables. Some relationships are +ve and some -ve. |
|
|
What is positive correlation? |
Two variable moving in the same direction. As one increases, so does the other. As one decreases, so does the other. |
|
|
What is negative correlation? |
Two variable moving in the opposite direction from one another. As one variable increases, the other variable decreases. |
|
|
What type of graph would be used for visualising correlation? |
Scatter graph Extra question - what shapes would the graphs look like with positive and negative correlation? Positive correlation: / Negative correlation: \ |
|
|
What does correlation not imply?.... 😉 |
CAUSATION... Duuuuh 😌 |
|
|
What range must the correlation coefficient be within? |
-1 and 1 |
|
|
What does it mean if the correlation coefficient is closer to 1 or -1 |
A stronger linear relationship. -1 or +1 = perfect linear relationships |
|
|
What are the two versions of the correlation coefficient and what type of tests are they? |
Pearson's Correlation Coefficient - parametric test Spearman's Rho - nonparametric test |
|
|
What is the Regression Line and how do you calculate linear regression? |
Regression Line = line of best fit Linear regression: Y = a + bX Y = value being predicted a = intercept of regression line - the point at which the trend line intersects the Y axis b = slope of regression line - divide a unit of distance along the Y axis by the corresponding unit along the x axis X = value of our predictor variable; the value we know |
|
|
What is the Regression Line and how do you calculate linear regression? |
Regression Line = line of best fit Linear regression: Y = a + bX Y = value being predicted a = intercept of regression line - the point at which the trend line intersects the Y axis b = slope of regression line - divide a unit of distance along the Y axis by the corresponding unit along the x axis X = value of our predictor variable; the value we know |
|
|
What is the coefficient of determination and what does it tell us? |
R^2 A measure of how well the regression line represents the data. If the regression line passes exactly through every point on the scatter plot, it would be able to explain all of the variation. It represents the percentage of the data that is the closest to the line of best fit. |
|
|
What is the Regression Line and how do you calculate linear regression? |
Regression Line = line of best fit Linear regression: Table- calculate XY, X^2 and Y^2, column totals (same as Pearson's) Y = a + bX Y = value being predicted a = intercept of regression line - the point at which the trend line intersects the Y axis b = slope of regression line - divide a unit of distance along the Y axis by the corresponding unit along the x axis X = value of our predictor variable; the value we know |
|
|
What is the coefficient of determination and what does it tell us? |
R^2 A measure of how well the regression line represents the data. If the regression line passes exactly through every point on the scatter plot, it would be able to explain all of the variation. It represents the percentage of the data that is the closest to the line of best fit. |
|
|
What are the steps in carrying out Pearson's correlation coefficient and Spearman's Rho? |
Calculate XY Calculate X^2 and Y^2 Calculate column totals Complete the equation and check significance R must be equal to, or larger than, the value in the table in order to be significant (ignoring the sign of r) |
|
|
What are the types of disgust? |
Core: revulsion at prospect of ingestion of offensive substance - distaste Animal nature: poor hygiene, inappropriate sex, injuries Interpersonal: contamination actual/imagined, moral unfairness |
|
|
Describe the Gape face? |
Raising upper lip and/or wrinkling of the nose |
|
|
What's the evolutionary advantage of disgust? |
Prevents ingestion/contact with diseases, toxins, etc. |
|
|
What's the evolutionary advantage of disgust? |
Prevents ingestion/contact with diseases, toxins, etc. |
|
|
What are social implications of disgust? |
|
|
|
What does Chi-Square goodness of fit test? |
Goodness of fit - examining one nominal variable and looking at how participants are allocated to different categories within that variable. Deals with frequencies of scores in each category. Categories must be mutually exclusive - independent. Looking at the distribution of observes frequencies compared to expected frequencies. The expected frequencies depend on the null hypothesis we are testing - there are two possible null hypotheses. |
|
|
What are the two possible null hypotheses when carrying out a Chi-Square goodness of fit? |
The population is evenly distributed across the categories (as you would expect by chance). There is no preference for one category over another.
The proportions in each category do not differ from a comparison population that has published expected frequencies |
|
|
For a Chi-Square, what must the alternative hypothesis be? |
A two tailed test |
|
|
What are the steps involved in carrying out a Chi-Square goodness of fit? |
1: calculate expected frequencies - an even distribution for one type of hypothesis. Total number in sample x proportion (i.e. If 20% = 20/100 = 0.2). 2: record observed frequencies 3: calculate Chi-Square O = observed E = expected For each cell work out last part of equation. Then add those values together. 4: check significance - Degrees of freedom = no. of categories -1 Value must be equal to or larger than the critical value 5. Interpret results |
|
|
What does Chi-Square test of association test? |
Two nominal variables and want to know if they're associated. Deals with frequencies of scores in each category comparing observed and expected frequencies |
|
|
What are the steps involved in carrying out a Chi-Square test of association? |
1: create contingency table 2: record observed frequencies 3: calculate expected frequency- (row total x column total)/TOTAL 4: calculate Chi-Square- O = observed E = expected For each cell work out second half of equation and then add all values together 5: check significance - Degrees of freedom = (R - 1) x (C - 1) R is number of rows. C is number of columns. Value must be equal to or larger than the critical value 5. Interpret results |
|
|
What is reliability? |
The extent to which a measurement is repeatable and consistent |
|
|
What are the three subsections of reliability? |
Internal consistency: within a test, people should respond consistently to all of the questions - reflects the extent to which items of a test measure various aspects of the same characteristic and nothing else. Inter-rater reliability: if you have two observers watching the same behaviour, their scores should agree with one another. Test-retest: If you give a test to a person more than once, they should get about the same score each time |
|
|
What is validity? |
The extent to which a test measures what it claims to measure. |
|
|
What are the three types of validity? |
Construct validity: you feel confident that the things you are manipulating and measuring in your study truly represent the ideas you have in mind. Relates to validity of your measure. External validity: you feel confident that you can generalise the results from your small sample group to people and settings outside your study Internal validity: you feel confident that you can make causal statements about what happened in your study. It is represented by the extent to which we can be sure that it is the IV that produces the effect that we see in the experiment. |
|
|
What are the 8 threats to internal validity? |
Selection Maturation - age/maturity and fatigue History - change occurring in the world between experiments Repeated testing - participants more chilled in the second one. Room may have been changed since the first. Instrumentation - refers to the collection of the data - researcher may have been in a better mood the second time. Experimental mortality - participants might die between studies or drop out due to heightened anxiety. Selection-Maturation - subject related variables and time related variables might interact. Experimenter bias - treating subjects differently, bias in selecting data to consider, the recording, interpretation and reporting of the data. |
|
|
What is a double blind design? |
Researcher and participants do not know if participant gets the control or experimental treatment. |
|
|
What are demand characteristics? |
A failure of the experimental situation that indicates to the participants: 1. How the researcher would ideally like them to respond in the study. 2. The true purpose of the study Subsequently participants alter how they respond. |
|
|
What is participant reactivity and what are 3 examples of it? |
Good participant role: participants respond in a way so that experimenter can confirm hypothesis Negative participant role: participants deliberately attempt to sabotage results and destroy the studies credibility Apprehensive participant role: participants concerned with how their responses will be evaluated; they feel anxious and fearful and this alters their response. |
|
|
What are the 8 parts of ethics in human research? |
Informed consent Motivation for being a subject Degree of risk or potential harm Right to withdraw Confidentiality - data protection act Protection of participants Debriefing Follow-up procedures to detect and mitigate any lasting adverse effects |
|
|
What are the viewpoints on animal research? What are the weaknesses of the viewpoints? |
Against: ethical status of animal=humans Weaknesses..... • unreasonable conclusions: removes limits to 'direct action' - indeed, requires it • equates animal use with holocaust, slavery • would you kill 1 dog to save all humans? Failure to recognise human awareness
Pros: Animals as tools, objects Weaknesses..... • Unreasonable conclusions: Would you kill all dogs all dogs for 1 human? Animal 'ownership' rather than 'stewardship' • Failure to recognise inherent worth. • The modelling paradox for human disease - to the extent animals are not like humans, then they are poor models. To the extent that they are like humans, then we should not use them |
