Cohen Ch 3 A Statistical Refresher

Front 1

Measurement Error

Back 1

influences on the test scores beyond what the test in intended to measure

Front 2

Alfred Binet viewed intelligence test data as:

Back 2

Orindal in nature

Front 3

which scale of measurement has a true zero point?

Back 3

ration scale

Front 4

which scale always involve measurement errors?

Back 4

continuous scales

Front 5

What are the 4 types of measurement scales?

Back 5

Nominal - scales that involve putting caterogories into mutually distinguish characters i.e male/female

Ordinal - uses the same as nominal but in addition they rank-order in numerical however these orders do not indicate units of measurement, has no absolute zero point

Interval - has features of Nominal and Ordinal however it has interval of groups i.e 20-30, 31-40 and so forth, has no absolute zero point

Ratio - Ratio scales are like interval scales except they have true zero points, i.e temperature

Front 6

3 measures of central tendency

Back 6

mean - median - mode

Front 7

3 measurements of variability?

Back 7

range, interquartile and semi-interquartile ranges and the standard deviation

Front 8

name some types of skewness?

Back 8

positive- leans towards the y axis and indicates a test being to hard or not being understood

Negative - lean away from the y axis and indicates a test as being to easy on test takers

Front 9

Name some Kurtosis?

Back 9

(def: steepness of a distribution in its center of a distribution)

Platykurtic - relatively flat in its curve

leptokurtic - relatively peaked in its curve

mesokurtic - somewhere in the middle

Front 10

Name some standard scores?

Back 10

z scores - conversion of a raw score into a number indicating how many standard deviation units the raw score is below or above the mean of distribution, T score - a scale that ranges from 5 standard deviations below the mean to 5 standard deviations above the mean, stanine - standard score with a mean of 5 and a standard deviation of approx. 2. dividing into 9 units (a term that was contraction of words standard and nine) takes on whole values from 1-9 which represents a range of performance that is half of a standard deviation in width, IQ scores - IQ scales are ordinally scaled. While one standard deviation is 15 points,

Front 11

what does normalising a distribution involve?

Back 11

stretching - the skewed curve into the shape of a normal curve thus creating a corresponding scale of standard scores

Front 12

explain the formula z = (X - M)/ SD

Back 12

x=raw scores, M= the representative sample and SD is the stand deviation

z scores have a mean of 0 and standard deviation of 1

Front 13

explain the formula ss = mean(X-M)/SD +

ex: mean = 15 Standard deviation = 100

Back 13

X = raw score, M = the representative sample mean, SD = standard deviation, ex: mean is 15 standard deviation is 100 raw score is 30 and z score is 3

Front 14

what is the break down of the normal curve?

Back 14

2.15%, 13.59%, 34.13% to the mean then the same again on the other side
in total middle 68.26% (adding 34.13% x2), next is 96.44% ( addings 2x 13.59 and 34.13) next 99.74% (adding all the breakdowns x 2)

Front 15

what is variance?

Back 15

equal to the arithmetic mean of the squares of the difference between the score in a distribution and their mean. (s2) s2=x2 divide by n.

Front 16

what is the standard deviation?

Back 16

measure of variability equal to the square root of the average squared deviations about the mean

Front 17

what are the standard score equivalents?

Back 17

z score mean 0 and SD 1
t score mean 50 SD 10
IQ score mean 100 SD 15

Front 18

what is the formula to convert z scores to standardised scores?

Back 18

standard score = 15(X_M)/SD + 100
standardised score = 15z + 100, therefore raw score of 30 and z score of 3 becomes standardised score of 145 (use a frequency distribution - normal curve mean being 100 then 15 gaps x 3 for z score = 145

Front 19

what is a linear transformation?

Back 19

a line that retains a direct numerical relationship to the original raw score.

Front 20

what is a nonlinear transformation?

Back 20

when the data under consideration is not normally distributed yet comparisons with a normal distribution needs to be made

Front 21

what is normalised standard scores?

Back 21

stretching the skewed curve into the shape of a normal curve and creating a corresponding scale of standard scores, this is done in order to compare scores on other tests.