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;
49 Cards in this Set
- Front
- Back
What are the four types of graphical representation? |
Bar charts Histograms Frequency polygons Scattergrams |
|
When should a bar chart be used? |
Used for nominal data (bars must never touch) Can also be used when plotting measures of central tendency with ordinal data |
|
When should a histogram be used? |
Used to show the distribution of scores measured along a continuous scale Used with ordinal or interval/ratio data, so bars must touch each other |
|
When should a frequency polygon be used? |
Used as a visual means of showing frequencies in continuous interval/ratio data Form of line graph - used with two sets of data |
|
When should a scattergram be used? |
Used for plotting correlations Shows a clear visual image of what correlation has been found |
|
How are inferential statistics used in psychological research? |
These allow researchers to draw conclusions about their research - They carry out stats tests to find the likelihood the results occurred by chance - Then they make a reasonable guess as to what the outcome of future exps will be - From these they decide whether to accept/reject null/alternative hypotheses |
|
How to researchers decide which hypothesis to accept? |
If changes in the dependent variable are due to chance, null is accepted If changes are due to IV; alternative is accepted |
|
What does the probability value represent? |
P value represents the probability of something happening by change |
|
What are the ranges of P? |
0-1 0 = no influence of chance 1 = complete influence of chance |
|
What are the different levels of significance? |
p=0.05 - means 5% likelihood that results occurred by chance
p=0.01 - means 1% likelihood |
|
When is a 1% significance level used? |
When theres a need to be highly certain that results aren't due to chance (e.g. medical tests) |
|
What is a type I error? |
When the null hypothesis is rejected/the alternative hypothesis is accepted when in fact the null was true - occurs when significance levels aren't strict enough |
|
What is a type II error? |
When the null hypothesis is accepted/alt is rejected when in fact it is false - occurs when significance levels are too strict |
|
What are the three questions used to identify which statistical test should be used? |
1. Are you looking for a difference or a relationship? 2. What type of experimental design (independent/repeated/matched)? 4. What type of data (nominal/ordinal/interval-ratio)? |
|
What are the three types of experimental design? |
Independent groups Repeated measures Matched pairs |
|
What is the independent groups design?
|
When the participants are exposed to only one condition within the study (different participants used for each) |
|
What is the repeated measures design? |
When the same participants are exposed to all the conditions within the study |
|
What is the matched pairs design? |
When participants in one condition are matched with a similar control in the other condition (matches based on features such as gender, age, cultural background, occupation etc) |
|
What are the three levels of measurement? |
Nominal Ordinal Interval/ratio |
|
What is nominal data? |
When the data falls into separate categories: each piece of data can only fall into one category |
|
What is ordinal data? |
When the data is ordered some way (e.g. height order, rank order etc) |
|
What is interval/ratio data? |
When data is measured in units of equal intervals and on a continuous scale - e.g., seconds, minutes, inches |
|
What are the rules of interval/ratio data when selecting the statistical test? |
You can either use Wilcoxon/Mann Whittney or T-T tests, but T-T is better |
|
What statistical tests are used for nominal data? |
Independent groups - Chi-Square (X) Repeated/Matched - Sign Test (S) |
|
What tests are used for either ordinal or interval/ratio data? |
Independent groups - Mann-Whittney (U) Repeated/Matched - Wilcoxon Signed Ranks (T) |
|
What tests are used for ONLY interval/ratio data? |
Independent groups - Unrelated T-T test Repeated/Matched - Related T-T test |
|
For which tests is the difference significant if the calculated statistics is LESS THAN OR EQUAL to the critical value? |
Mann-Whittney/Wilcoxon/Sign tests |
|
For which tests is the difference significant if the calculated statistics is MORE THAN OR EQUAL to the critical value? |
Chi-Square/Spearman/T-T tests |
|
What 4 things do you need to know in order to find the correct critical value? |
1. Number of participants 2. Whether a one-tailed or two-tailed test was used 3. Significance level used 4. Degrees of freedom (Chi-Square and T-T tests) |
|
What is a correlational study? |
Testing to see if there is a relationship between two variables Relationship NOT difference |
|
What is a Spearman's test/rho? |
Spearman's rho is the statistical test/value used when trying to identify whether a relationship is significant in a correlational study |
|
What are the requirements for a Spearman's test? |
1. Looking for a relationship not difference 2. Data is either ordinal or interval/ratio 3. Data is in the form of related pairs (repeated measures) |
|
What is the correlation coefficient? |
The calculated value from a Spearman's test |
|
When is a Pearson's test used |
The alternative test for non-related pairs Data must be interval/ratio and drawn from a normal distribution |
|
What is qualitative data? |
Emphasis on quality Focus of meaning For example, case studies and unstructured interviews Can be turned into quantitative or be purely qualitative |
|
What is quantitative data? |
Emphasis on quantity Focus on numerical forms Experiments and content analyses |
|
What is coding and how is it used? |
Its the process of identifying categories/themes/phrases/keywords that may be found in any set of data
Coding aims to understand the meaning of the data obtained Group-like statements can be put together (categorisation) |
|
What is the top-down approach to coding? |
Codes used are taken from an existing theory; codes therefore decided before observation/interview takes place |
|
What is the bottoms-up approach to coding? |
Categories are formed as a result of the data obtained - codes aren't decided prior to observation/interview but after |
|
When is the bottoms-up approach usually used? |
When the area of investigation is less researched |
|
What is triangulation? |
It assesses the validity of qualitative data by comparing results from a variety of different studies of the same topic/person - if the results differ then further research is needed |
|
What is objectivity in psychological research? |
When the research remains detached in the research process They don't let personal feelings affect results Considered scientific Produces quantitative data |
|
What is an example of objectivity? |
Methods considered objective are by using alternative interviewer to themselves to conduct interviews to prevent researcher bias |
|
What is subjectivity in psychological research? |
When the researcher may not remain completely detached from the process A degree of interpretation involved in the process thus considered less scientific Produces qualitative data |
|
What is an example of subjectivity? |
Methods considered to be subjective are observing and recording down violent behaviour in children in a playground setting as differing opinions on what is considered 'violent' |
|
What are the advantages of producing qualitative data? |
A) Data gives us richness of detail, and insights into the way people feel - This isn't possible using more objective methods and helps us to represent true complexities of human behaviour |
|
What are the disadvantages of producing qualitative data? |
A) Its harder to detect patterns as less participants are focused on - therefore more difficult to draw conclusions B) Its subjective meaning interpretation of results may be biased and affected by researcher's own expectations and beliefs - there could be other interpretations, making the method much less scientific |
|
What are the advantages of producing quantitative data |
A) Its easier to analyse because its in numbers - allows us to carry out stats tests producing scientific results B) Gives us objective data which isn't open to interpretation - means psychologists can check reliability of results |
|
What are the disadvantages of producing quantitative data? |
A) It oversimplifies reality and human experience - We're more complex than numbered results in one study |