Research Methods 1&2 Flashcards Preview

Psychology > Research Methods 1&2 > Flashcards

Flashcards in Research Methods 1&2 Deck (48):
1

What does the PPPAC Cycle of research stand for?

->Problem-> Plan -> Data -> Analysis -> Conclusion ->

2

Why is the research question important in statistical analysis?

The research question drives the research - it is the reason for the research i.e it gives the research it's purpose and context.

3

What is the difference between descriptive and psychological interpretations?

Descriptive = describe relationships e.g men have higher self-esteem on average than women

Psychological = explanatory/drawing conclusions e.g why do men have higher self-esteem on average than women.

4

What is a hypothesis?
What is the difference between a theoretical hypothesis and a research hypothesis?

Hypothesis = a plausible answer to a research question
Theoretical hypothesis = Regards causal links between constructs
Research hypothesis = regards specific statements about variables i.e expected observations.

5

What are the four different measurement formats?

Nominal, Ordinal, Interval, Ratio

6

Describe nominal measurement formats

Categorical (non-continuious), no 'rank-order' e.g gender: male or female

7

Describe Ordinal measurement formats

Similar to nominal - both are categorical. However Ordinal categories are in rank order i.e some categories represent more of the variable than another. e.g, first to last place

The difference between the categories DOES NOT have to be equal

8

Describe Interval and ratio measurement formats

Categories in rank order where the difference between categories are EQUAL.

e.g age in years (NOT under 18, 18-25, 26-39 <- this is ordinal)

Ratios differ from interval variables in that ratio variables have a true zero i.e a score of 0 indicates that a person has none of that property.

9

What is a categorical variable?
Which two of the four measurement formats are categorical?

Variables where the different levels represent different categories but provide no information about the differences between levels (or categories). i.e not meaningful as a number. E.g religious affiliation or gender.

Nominal and Ordinal

10

What is a continuous variable?
Which two of the four measurement formats are categorical?

Variables where the different levels of a variable have a definite order and the differences between levels are known to be equal. i.e meaningful as a number. (e.g counting up test results and age in years)

Interval and ratio

11

What are 'descriptive statistics'?

aka 'summary statistics' describe data in a simpler form, merely listing all the scores obtained on all of the variables.

12

What are the three main types of descriptive statistics?

Univariate, bivariate, multivariate

13

What are univariate statistics?

Describe the distribution of scores on a single variable. (e.g what sort of scores do we see, is a particular score more common, how are the scores distributed)

14

What are bivariate (and multivariate) statistics?

describe relationships (covariation) between two (or more than two) variables. e.g whether scores on one variable tell us something about scores on another. Looking for covariation (a relationship) is the first sign that there could be causation.

15

What are inferential statistics?

compare descriptive statistics with what we might expect by chance alone (Subject to certain assumptions). Based on mathematical probability theory and are statements about the probability of observed results. I.E looking at whether the relationship we find are significant in some way - or just a product of chance.

16

What type of statistics do you run on your data set first and what second? why?

Descriptive statistics, than inferential statistics.

Because you should evaluate your hypothesis first using descriptive statistics to make a tentative evaluation and than refine it further via inferential statistics.

17

What is raw data?

The untouched data from research that has no been summaries/organised with descriptive statistics.

e.g scores from a first year psyc test = 5,6,7,7,8,8,8,9,9,10,10,11,11,11,12,12,12,12,12 etc

18

What is a frequency/frequency count? (Descriptive statistics)
What are percentages?

A frequency count is the number of time a certain score/categorie occurs e.g there were 11 males and 39 females. These are only meaningful if you know the sample size.

This is often converted into a percentage by dividing each frequency by the total number of scores (e.g 50 participants) and then multiplying by 100 (in this case = 22% males and 78% females)


Good for categorical variables.

19

What are the two types of descriptive statistics used for summarising data from continuous variables?

Central tendency (e.g the mean, median, mode) and variability (e.g the range, standard deviation)

20

What is the standard deviation?

A measurement of the average (mean) variability in scores from the mean.

21

What is the optimal descriptive statistics for categorical and continuous variables?

Categorical = percentages
Continuous = central tendency (e.g mean) accompanied by index of variability (e.g SD)

22

What is the best descriptive statistic graphical representation for percentages (categorical variable)?

Bar chart.

23

What is the best descriptive statistic graphical representation for continuous variables?

Histogram

24

What is the interquartile range (IQR)?

IQR is the range of the middle 50% of scores. (thus it is not influenced by the extreme ends)

The difference between the scores at the 3rd and 1st quartiles. The third quartile is the median of the highest fifty percent of scores, while the first quartile is the median of the lowest fifty percent of scores. in otherwords 25% of scores fall below the first quartile and 50% of scores falls below the median and 75% of scores fall below the third quartile.

25

What is a 'unimodal' distribution?

A distribution (of scores) with one main peak.

26

What is a 'bimodal' or 'multimodal' distribution?

A distribution (of scores) with two, or more, main peaks.

27

Is the mean a good indicator of central tendency for bimodal and multimodal distributions?

No. For such distributions, the central tendency is best described by noting all of the modal categories (i.e., score intervals where multiple peaks occur).

28

What is a skewed distribution?
What is a symmetrical distribution?

Skewed = where scores tend to predominate at one end of the distribution.
Symmetrical = when the left and right hand halves of the distribution are mirror images of each other (or, at least, are approximately so).

29

When a distribution is skewed such that there are more low scores than high scores, is it positively or negatively skewed? (i.e the peak is on the left)

Positively skewed.

30

When a distribution is skewed such that there are more high scores than low scores, is it positively or negatively skewed? (i.e the peak is on the right)

Negatively skewed.

31

What is the 'normal distribution'?

A unimodal symmetric bell-shaped distribution (iwht a particular formula). Common and important for inferential statistics.

32

What is the 68-95-99.7 rule for SD in a normal distribution?

68% of observations fall within 1SD of the mean
95% of observations fall within 2SD of the mean
99.7% of observations fall within 3SD of the mean

33

Describe the SD properties of a normal distributions.
(i.e what % of observations fall within 1, 2 and 3 SD of the mean)

the 68-95-99.7 rule

68% of observations fall within 1SD of the mean
95% of observations fall within 2SD of the mean
99.7% of observations fall within 3SD of the mean

34

What does a standard score or Z-score tell us?

How many SD a specific score is away from the mean.

A negative Z score = below the mean
A positive Z score = above the mean

35

In bivariate descriptive statistics, what are the four possible combinations of scales of measurement (given dependent v. independent variable)?

2x categorical variables (IV x DV)
1x categorical (IV) 1x continuous (DV)
1x categorical (DV) 1x continuous (IV)
2x continuous variables

36

What is the best descriptive statistic strategy for a bivariate relationship with a categorical IV and a continuous DV?

Difference between means (and SD).

i.e comparing the mean scores of the (continuous) dependent variable between each level of the independent variable.


37

How do we get the Cohens D effect size?

and what is cohens d a measure of?

d = (Mean1 - Mean 2)/ SDpooled

for a conservative estimation, use the larger of the two SD.

cohens d is a measure of how many SD there is between 2 means.

38

What is the best descriptive statistic strategy for a bivariate relationship with a categorical IV and a categorical DV?

Difference between percentages.

i.e comparing the percentages for each level of the dependent variable across each level of the independent variable.

39

What effect size measure is used when comparing means? (Categorical IV, continuous DV)

Cohens d

(i.e how many SD there is between 2 means.)

40

What effect size measure is used when comparing percentages? (Categorical IV, categorical DV)

Cohens w

OR a difference in percentage by 30% or greater = medium, and thus important, sized effect.

41

What is a contingency table?

Used to present bivariate descriptive data regarding categorical IV and DV. Presents the frequencies and percentages for all groups.

42

What is the best descriptive statistic strategy for a bivariate relationship with a continuous IV and a continuous DV?

what is the optimal graphical representation?

Correlation coefficient
regression coefficient

Graphical representation = scatter plot
(regression)

43

On a scatter plot on which axis are the IV and DV?

IV = x-axis (i.e horizontal)
DV = y-axis (i.e vertical)

44

What are the two important features of a scatter plot?

Strength and Direction

45

On a scatter plot, what is a positive and a negative relationship?

positive relationship: people with high scores on one variable are likely to have high scores on the other.
negative relationship: people with high scores on one variable are likely to have low scores on the other.

46

what is the line of best fit on a scatter plot?

A line drawn through the data points so that it is on average closest to all the points in the graph.

47

In regression coefficients, what does a and b stand for?

a = value of Y when X is at 0 (i.e where the regression line crosses/intercepts the Y axis)

b = The slope of the regression line, the predicted change in the dependant variable for every 1 increase on the independent variable. (not helpful for determining strength of relationship - must be converted to standard score - i.e correlation coefficient)

48

what size effect would: r = .10, .30, .50 (either - or +); be considered to be?

Small, Medium, Large