arithmetic mean

a measure of central tendency

derived by calculating an average of all scores in a distribution

also called the mean
p.89
average deviation

a measure of variability

derived by

summing the absolute value of all the scores in a distribution and

dividing by the total number of scores
 summing the absolute value of all the scores in a distribution and
 dividing by the total number of scores
p. 94
bar graph

a graphic illustration of data

numbers indicative of frequency are set on the vertical axis

categories are set on the horizontal axis

the rectangle bars that describe the data are typically noncontigous
p. 85
bimodal distribution

a distribution where the central tendency consists of

two scores

occurring an equal number of times

and are the most frequently occurring scores in the distribution
 two scores
 occurring an equal number of times
 and are the most frequently occurring scores in the distribution
p. 91
bivariate distribution

a graphic representation of correlation

accomplished by the simple graphing of the coordinate points for values of

the Xvariable and

the Yvariable

also known as a

scatterplot

scatter diagram

scattergram
 the Xvariable and
 the Yvariable
 scatterplot
 scatter diagram
 scattergram
p. 111
coefficient of correlation

symbolised by r

the correlation coefficient is an index of the strength of the linear relationship between two continous variables

expressed as a number

can range from 1 to +1

most frequently used statistic to calculate is the Pearson r
 can range from 1 to +1
p.106
coefficient of determination

a value indicating how much variance is shared by two variables

this value is obtained by

squaring the obtained correlation coefficient

multiplying by 100 and

expressing the result as a percentage

this indicates the amount of variance accounted for by the correlation coefficient

i.e., "total variance explained"
 squaring the obtained correlation coefficient
 multiplying by 100 and
 expressing the result as a percentage
 i.e., "total variance explained"
p.109
correlation

an expression of the degree and direction of relationship (correspondence) between two variables

where each variable is continuous in nature
p.106116
curvilinearity

the degree to which a graph or scatterplot is characterized by curvature
p.111
distribution

in a psychometric context

a set of test scores arrayed for recording or study
p.83
dynamometer

an instrument used to measure the strength of hand grip
p. 82
effect size

a statistic used to express

the strength of a relationship or

the magnitude of the differences in data

in metaanalysis  this statistic is most typically a correlation coefficient
 the magnitude of the differences in data
p.115
error

all of the factors (other than what a test purports to measure) that contribute to scores on a test

error is a variable that exists on all testing and assessment
p.7879
frequency distribution

a tabular listing of

scores

along with the number of times each score occurred
 scores
 along with the number of times each score occurred
p.8385
frequency polygon

a graphic illustration of data

frequency numbers set on the vertical axis

test scores or categories are set on the horizontal axis

data is described by a continous line connecting all of the points where the test scores or categories meet frequencies
p.85,86
graph

a diagram or chart

composed of lines, points, bars, or other symbols that describe and illustrate data
p.85
grouped frequency distribution

a tabular summary of test scores

the test scores are grouped by intervals

also referred to as class intervals
p. 8485
histogram

a graph with vertical lines drawn at the true limits of each test score (or class interval)

these lines form a contigous rectangle
p.85,86
interquartile range

an ordinal statistic of variability

equal to the difference between the third & first quartile points in a distribution that has been divided into quartiles
p.94
interval scale

a system of measurement

all items are rankordered into equal intervals

every unit on the scale is equal to every other

there is no absolute zero point

this precludes mathematical operations on the data
 this precludes mathematical operations on the data
p.81
kurtosis

an indication of the nature of the steepness of the centre of the distribution

i.e., peaked vs flat
 i.e., peaked vs flat
p.9798
leptokurtic

a description of the kurtosis of a distribution that is relatively peaked at its centre
linear transformation

in psychometrics

a process of changing a score such that

the new score has a direct numerical relationship to the original score

the magnitude of the difference between the new score & other scores on the scale parallels the magnitude of differences on the scale from which it was derived

contrast with nonlinear transformation
 the new score has a direct numerical relationship to the original score
 the magnitude of the difference between the new score & other scores on the scale parallels the magnitude of differences on the scale from which it was derived
p.104
mean

a measure of central tendency

derived by calculating an average of all scores in a distribution

also called an arithmetic mean
p.89
measurement

assigning numbers or symbols to characteristics of people or objects

according to rules
 according to rules
p.7879
measure of central tendency

one of three statistics indicating the average or middlemost score between the extreme scores in a distribution

mean  ratio level

median  ordinal (takes the order of scores into account)

mode  nominal
p.89
measure of variability

a statistic indicating how scores in a distribution are scattered or dispersed

common measures of variability are

range

standard deviation

variance
 range
 standard deviation
 variance
p.92
median

a measure of central tendency

derived by identifying the middlemost score in a distribution
p.8991
mesokurtic

a description of the kurtosis of a distribution

that is neither extremely peaked nor flat in its centre
 that is neither extremely peaked nor flat in its centre
p.97
metaanalysis

a family of techniques

used to statistically combine information across studies

to produce single estimates of the statistics being studied
p.115
mode

a measureof central tendency

derived by identifying the most frequently occurring score in a distribution
p.89, 9192
negative skew

when relatively few scores fall at the lower end of the distribution

negatively skewed exmination results may indicate the test was too easy

some more difficult questions would better discriminate between scores at the higher end
 some more difficult questions would better discriminate between scores at the higher end
p.97
nominal scale

a system of measurement

all things being measured are classified or categorised, based on one or more distinguishing characteristics

placed into mutually exclusive & exhaustive categories
p.7980
nonlinear transformation

in psychometrics

a process of changing a score such that

the new score does not necessarily have a direct numerical relationship to the original score, and

the magnitude of the diffferences between the new score & the other scores on the scale, may not necessarily parallel the magnitude of differences of those from the original scale

contrast with linear transformation
 the new score does not necessarily have a direct numerical relationship to the original score, and
 the magnitude of the diffferences between the new score & the other scores on the scale, may not necessarily parallel the magnitude of differences of those from the original scale
p.104
normal curve

bellshaped, smooth, mathematically defined curve

highest at the centre & gradually tapered on both sides  approaching but never touching the horizontal axis
p.85, 87, 98
normalized standard score scale

conceptually, the end product of "stretching" a skewed distribution into the shape of a normal curve

usually through a nonlinear transformation
p.104106
normalizing a distribution

a statistical correction applied to distributions meeting certain criteria

for the purpose of approximating a normal distribution

thus making the data more readily comprehensible or manipulable
 for the purpose of approximating a normal distribution
 thus making the data more readily comprehensible or manipulable
p.104
ordinal scale

a system of measurement , where all things being measured can be rankordered

the rankordering implies nothing about how much greater one ranking is than another

no absolute zero point on the scale

most scales in psychology & education are ordinal
p.8081
outlier

an extremely atypical plot point in a scatterplot

any extremely atypical finding in research
p.111
Pearson r

a widely used statistic for obtaining an index of the relationship between two variables

when that relationship is linear and

the two correlated variables are continuous (i.e., can theoretically take any value)

also known as

the Pearson coefficient of productmoment correlation and

the Pearson correlation coefficient
 when that relationship is linear and
 the two correlated variables are continuous (i.e., can theoretically take any value)
 the Pearson coefficient of productmoment correlation and
 the Pearson correlation coefficient
platykurtic

a description of the kurtosis of a distribution that is relatively flat in its centre
p.97
positive skew

when relatively few of the scores fall at the high end of the distribution

positively skewed examination results may indicate the test was too difficult

some easier questions would better discriminate at the lower end of the distribution
 positively skewed examination results may indicate the test was too difficult
 some easier questions would better discriminate at the lower end of the distribution
p.97
quartile

one of the three dividing points between the four quarters of a distribution

each typically labelled

Q1

Q2

Q3

Q4
 Q1
 Q2
 Q3
 Q4
p.93
range

a descriptive statistic of variability

derived by calculating the difference between the highest & lowest scores in a distribution
p.93
rankorder/rankdifference correlation coefficient

an index of correlation

statistic of choice when

sample size is small, and

both sets of measurement are ordinal

also referred to as the Spearman's rho
 sample size is small, and
 both sets of measurement are ordinal
p.110111
ratio scale

a system of measurement where all things being measured can be rankordered

the rankordering does imply something about exactly how much greater one ranking is than another

equal intervals exist between each number on the scale

all mathematical operations can be performed meaningfully

because a true zero point exists

few scales in psychology & education use ratio scales
 because a true zero point exists
p.8183
raw score

a straight forward, unmodified accounting of performance

usually numerical

typically used for evaluation or diagnosis
scale

1) a system of ordered numerical or verbal descriptors

usually occurring at fixed intervals

used as a reference standard in measurement

2) a set of numbers or other symbols whose properties model empirical properties of the objects or traits to which numbers or other symbols are assigned
 usually occurring at fixed intervals
 used as a reference standard in measurement
scatter diagram

a graphic description of correlation

achieved by graphing the coordinate points for the two variables

also referred to as a

scatterplot

scattergram, or

bivariate distribution
 scatterplot
 scattergram, or
 bivariate distribution
p.111
scattergram

a graphic description of correlation

achieved by graphing the coordinate points for the two variables

also referred to as

scatterplot

scatter diagram or

bivariate distribution
 scatterplot
 scatter diagram or
 bivariate distribution
p.111
scatterplot

a graphic description of correlation

achieved by graphing the coordinate points for the two variables

also referred to as

scatter diagram

scattergram, or

bivariate distribution
 scatter diagram
 scattergram, or
 bivariate distribution
p.111
semiinterquartile range

a measure of variability

equal to the

interquartile range divided by two
 interquartile range divided by two
p.94
skewness

an indication of the nature & extent to which symmetry is absent in a distribution

a distribution is said to be

skewed positively when relatively few scores fall at the positive end and

skewed negatively when relatively few scores fall at the negative end
 skewed positively when relatively few scores fall at the positive end and
 skewed negatively when relatively few scores fall at the negative end
p.9697
Spearman's rho

an index of correlation

statistic of choice when

sample size is small, and

both sets of measurement are ordinal

also referred to as the

rankorder correlation coefficient, and

rankdifference correlation coefficient
 sample size is small, and
 both sets of measurement are ordinal
 rankorder correlation coefficient, and
 rankdifference correlation coefficient
p.110111
standard deviation

a measure of variability

equal to the square root of the averaged squared deviations about the mean

also equal to the square root of the variance
p.9496
standard score

a raw score that has been converted from one scale into another

the new scale has

arbitrarily set M & SD

is more widely used & readily interpretable

examples of standard scores are

z scores

T scores
 arbitrarily set M & SD
 is more widely used & readily interpretable
 z scores
 T scores
stanine

a standard score derived from a scale with

a mean of 5 and

a standard deviation of approx. 2
 a mean of 5 and
 a standard deviation of approx. 2
p.103104
T score

a standard score

calculated using a scale with

a mean set at 50 and

a standard deviation set at 10

used by the developers of the MMPI

named for Thorndike
 a mean set at 50 and
 a standard deviation set at 10
p.103, 429
tail

the area on the normal curve between

2 and 3 standard deviations above the mean, and

2 and 3 standard deviations below the mean

a normal curve has two tails
 2 and 3 standard deviations above the mean, and
 2 and 3 standard deviations below the mean
p.101
variability

an indication of how scores in a disrtibution are scattered or dispersed
p.9296
variance

a measure of variability

equal to the

arithmetic mean of the squares of the differences

between the scores in a distribution

and their mean
 arithmetic mean of the squares of the differences
 between the scores in a distribution
 and their mean
p.95, 146
z score

a standard score

derived by calculating the difference between

a particular raw score & the mean

and then dividing by the standard deviation

a z score expresses a score in terms of the number of standard deviation units the raw score is below or above the mean of the distribution
 a particular raw score & the mean
 and then dividing by the standard deviation
p.102103