statistics

Question 1

criteria for using parametric tests

Answer 1

populations drawn should be normally distributed
variances of population should be approxamiately equal
at least interval level data
no extreme scores

Question 2

what is the rule for a test to be significant

Answer 2

if the test has the letter “r” in it, for significance the observed value needs to be greater than critical value

Question 3

what is nominal data

Answer 3

data that can be put it into categories

Question 4

strength of nominal data

Answer 4

Easy to generate from closed questions; large amounts of questions can be collected quickly; increasing reliability

Question 5

weakness of nominal data

Answer 5

only mode can be used as a measure of spread
doesn’t provide any explanations for data

Question 6

what is ordinal data

Answer 6

results are able to be ranked in order
but the difference between each score is not known and does not have to be equal
e.g. test scores

Question 7

strength of ordinal data

Answer 7

can find median and mode
provides more info than nominal as indicates values on a scale

Question 8

weakness of ordinal data

Answer 8

gaps between points are unknown so is hard to compare and can be interpreted differently

Question 9

what is interval data

Answer 9

most precise level of data
scale of equal or known units with equal distances
e.g. temperature, time in seconds

Question 10

strength of interval data

Answer 10

most informative
points are linear with equal gaps so mean, modian and mode can all be calculated
standard deviation can be calculated to analyse dispersion
most reliable and scientific

Question 11

weakness of interval data

Answer 11

do not gain any answers on why , lack of explanation

Question 12

how to calculate mean
strengths and wekanesses

Answer 12

add all data togheter then divide by number of pieces of data
strength- includes all pieces of data so is an accurate representation
weakness- extreme scores can distort the value, so is not an appropriate measure of central tendency if data set is skewed.

Question 13

how to calculate median
strenghts and weaknesses

Answer 13

all numbers in order, is middle number, if there are 2 numbers in middle add together and divide by 2.

one strength is extreme values do not distort the value
is a strong measure of central tendency even if data set is skewed
one weakness is it could be time consuming and difficult to calcuate with a large set, is less representative as does not depend on all pieces of data

Question 14

how to calculate mode
strenghts and weaknesses

Answer 14

mode is the one that appears the most

strenght- can use when data is quantatitive or qualitative, so always allows for analysis for most frequent category

weakness- does not accurately reflect the data set
only works out most popular, so is useless

Question 15

what is a measure of central tendency

Answer 15

single data point which summarises whole data srt
for example mode, median, mean
is a summary measure

Question 16

what is a measure of dispersion

Answer 16

measures level of spread or variability in data set
for example range, interquartile range and standard deviation

Question 17

what is discrete data

Answer 17

can be counted
distinct and speerate pieces of data
for example number of CD’s sold

Question 18

what is continous data

Answer 18

has infinite value
e.g. height, time, age

Question 19

characterisitcs of bar chart

Answer 19

discrete data
not touching colums
used for comparing categoies
dependent variable on Y axis
independent variable on X axis
has equal intervals

Question 20

characterisitcs for histograms
including equation

Answer 20

continious data
columns touch
areas of columns show frequency
frequency density is on Yaxis
continious scale on X axis

frequency density= frequency/class width

Question 21

characterisitcs for pie charts

Answer 21

discrete data
shows contribution to overall total
shown as %
represented as an angle

Question 22

characterisitcs for line graphs

Answer 22

continious data
compares 2 or more conditions
continious scale on X axis
frequency on Y axis
data points connected with straight line

Question 23

characterisitcs of scatter diagrams

Answer 23

continious data
measures relationship between 2 variables
one variable on X axis, the other on Y axis
shows correlation

Question 24

what is range
strenghts and weaknnesses

Answer 24

range is highest answer take lowest

is easy and quick to calculate
but
affected by extreme values
doesn;t take into account distribution around mean

Question 25

what is variance how to work it out strengths and weaknesses

Answer 25

calculates difference between each piece of data and mean, the larger the standard deviation, the more spread out data is. find the mean, find the difference between piece of data and mean, square each difference, sum the squares, and divide by the number of data points minus one. strengths-takes all scores into account, more representative weaknesses- may hide some characterisitcs of data, less useful on its own, often need standard deviation

Question 26

what is standard deviation strenghts and weaknesses

Answer 26

tells us average number that differs from mean is the square root of variance strenghts- most precise measure of dispersion, all values considered returns to units same figure as mean making it easier to judge the smaller the SD, the more similar it is to mean

Question 27

draw negatively skewed normal skewed and positively skewed labelling mean, median, mode

Question 28

when do normal distribution occur

Answer 28

when variables distributed so most of scores are clustered around mean, median and mode 50% of scores to left, 50% to right of mean

Question 29

when does positively skewed distribution occur

Answer 29

most scores bunched to left mode is to left of mean as mean is affected by anamolies

Question 30

when does negatively skewed distribution occur

Answer 30

most scores are bunched to the right mode is to right of mean

Question 31

type one error definition

Answer 31

says results are significant when theyre not

Question 32

type two error defintion

Answer 32

says results are not significant when they are

Question 33

when would you use chi square test

Answer 33

independent measures nominal data test of difference

Question 34

when would you use mann whitney test

Answer 34

independent measures ordinal/interval data test of difference

Question 35

when would you use wilcoxon test

Answer 35

ordinal/ interval repeated measures test of differences

Question 36

when would use binomial test

Answer 36

repeated measures nominal test test of differences

Question 37

when would you use spearmans rho

Answer 37

at least ordinal data exploring relatioship between co variables correlational design

Question 38

steps for mann whitney

Answer 38

rank both sets of data together place them back into table find overall ranking for each category either r1 or r2 use smallest total (r1 r r2) in the equation u1=r1-n(n+1)/2 n= number of particpants to be significant. observed U value has to be equal or less than critical value at p 0.05 significance level

Question 39

steps for wilcoxon

Answer 39

find differences between data sets rank the differences, if difference is 0 ignore count how many positive and negative there are add together the differences of less frequent sign n= number of differences (not including 0s) to be significant oberved value has to be equal or less than crtical at p 0.05 significance

Question 40

steps for chi square

Answer 40

(piece of data already in cell is observed) find totals for each column find expected frequencies for each cell using row total x column total/ overall total use formula (observed-expected)2/expected for each cell finally add all numbers for each cell together calculate degree of freedom using (number of rows - 1) x (number of columns - 1) observed chi square value needs to be more than critical value to be significant

Question 41

steps for binomial test

Answer 41

label flow of direction if same answer in both categories ignore count number of each direction of flow smallest number of direction of flow is observed n= number of pps whos scored was use (don't include the ignore) to be significant, observed value has be smaller or equal to critical binomial

Question 42

steps of spearmans rho

Answer 42

rank each column individally find differences between ranks square differences this is sgima d 2 google equation cant inset it lol x n= number of participants do not ignore zeros score will be between 0 and 1 for significance, observed value needs to be more than spearmans rho crtical value