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;
21 Cards in this Set
- Front
- Back
Measurement Error
|
influences on the test scores beyond what the test in intended to measure
|
|
Alfred Binet viewed intelligence test data as:
|
Orindal in nature
|
|
which scale of measurement has a true zero point?
|
ration scale
|
|
which scale always involve measurement errors?
|
continuous scales
|
|
What are the 4 types of measurement scales?
|
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 |
|
3 measures of central tendency
|
mean - median - mode
|
|
3 measurements of variability?
|
range, interquartile and semi-interquartile ranges and the standard deviation
|
|
name some types of skewness?
|
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 |
|
Name some Kurtosis?
|
(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 |
|
Name some standard scores?
|
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,
|
|
what does normalising a distribution involve?
|
stretching - the skewed curve into the shape of a normal curve thus creating a corresponding scale of standard scores
|
|
explain the formula z = (X - M)/ SD
|
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 |
|
explain the formula ss = mean(X-M)/SD +
ex: mean = 15 Standard deviation = 100 |
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
|
|
what is the break down of the normal curve?
|
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) |
|
what is variance?
|
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.
|
|
what is the standard deviation?
|
measure of variability equal to the square root of the average squared deviations about the mean
|
|
what are the standard score equivalents?
|
z score mean 0 and SD 1
t score mean 50 SD 10 IQ score mean 100 SD 15 |
|
what is the formula to convert z scores to standardised scores?
|
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 |
|
what is a linear transformation?
|
a line that retains a direct numerical relationship to the original raw score.
|
|
what is a nonlinear transformation?
|
when the data under consideration is not normally distributed yet comparisons with a normal distribution needs to be made
|
|
what is normalised standard scores?
|
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.
|