RM and stats defs

Question 1

raw numbers

Answer 1

unrefined data (refined to make descriptive statistics)

Question 2

central tendency

Answer 2

averages: mean (midpoint), median (midpoint), mode (typical)

Question 3

dispersion

Answer 3

how spread out around average: range, standard deviation
when 2 number sets have the same average but different level of variation.
- shows us how focused/ representative the average is

Question 4

CT of Mean

Answer 4

• easily distorted by outliers
• can give silly answers for binary stats, e.g, average person has 1.78… legs bc some only have 1
  use: usually/default, unless many outliers

Question 5

CT of Median

Answer 5

• not distorted like mean as always lies in middle
• non holistic view
  use: when too many outliers for mean

Question 6

CT of Mode

Answer 6

• not as often used - does not describe the middle

use: when collecting frequencies, non numerical data

Question 7

CT of Range

Answer 7

• clear description of diversion

- v vulnerable to outliers g.g, extreme smallest/largest while majority are close

Question 8

CT of Standard Deviation

Answer 8

(average deviation from average/ct)

SD=root(total deviation^2/N–1)

Question 9

how to calculate standard deviation

Answer 9

1) calculate deviation of each stat from mean (stat–mean)
2) square each deviation and find total
3) divide total by number of scores–1 (deg.s of freedom)
4) square root answer

Question 10

Variance

Answer 10

standard deviation^2
used to also measure distribution of results

1) calculate deviation of each stat from mean (stat–mean)
2) square each deviation and find total
3) divide total by number of scores–1 (deg.s of freedom)

Question 11

standard error of the mean

Answer 11

used to test how representative the mean is (uncertainty)

• number of people tested
• how consistent they are (similar)
• low value is good

Question 12

types of data

Answer 12

Categorical - “Nominal”- categories with no intrinsic ranking e.g, female/male, names
Discrete - “Ordinal” - with intrinsic ranking - e.g, rating scales
Continuous - “Scale” - (number scale - interval (integers) or ratio (decimals) - scalar variable where axis represents a metric, e.g, height, test scores

Question 13

central limit theory

Answer 13

theory that the averages (means) of data collected from samples will form a normal distribution curve around the original mean of the results

Question 14

p value

Answer 14

if >0.05 results oppose research hypothesis enough to consider null hypothesis is correct

Question 15

one vs two tailed hypothesis

Answer 15

one tailed: predicts that the independent variable will have an effect on the dependent variable, but the direction of the effect is not specified.
2 tailed: predicts either +ve or -ve impact of iv on dv

Question 16

chi squared test - function and when used?

Answer 16

To test for independence of 2 variables - independent or related?

• a non-parametric test - used for not normally distributed data
• nominal scale - e.g, female vs males, handwritten vs typed
• mode used