Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Test Theory > Chapter 2: Essential statistics for testing > Flashcards

Chapter 2: Essential statistics for testing Flashcards

1
Q

measurement

A

application of rules for assigning numbers to objects or events
- systematic analysis, categorization, and quantification of observable phenomena

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Variable

A

anything that varies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Constant

A

anything that is held constant

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

discrete variables

A

variables with a finite range of values (countable range of values)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

dichotomous variables

A

discrete variables that can assume only two values

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

polytomous variables

A

discrete variables that can assume more than two values

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

continuos variables

A

have high infinite ranges and cannot really be counted (Time, distance, temperature)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Nominal (scale type)

A

numbers are used instad of words (identity or equality)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

ordinal (scale type)

A

numbers are used to designate an orderly series (identity + rank order)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

interval (scale type)

A

equal intervals between units but not true zero (identity + rank oder + equality of units)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

ratio (scale type)

A

zero means none of whatever is measured, all arithmetic operations are possible and meaningful (identity + rank oder + equality of units + additivity)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

categorial data

A

data that derive from assigning people, objects, or events to particular categories or classes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

property of “identity”

A

all memebrs of a category must be assigned the same number and that no two categories may share the same number

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

rank order

A

rank order numbers convey precise meaning in terms of positions BUT they carry no info with regard to the distance between positions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

percentile rank (PR) scores

A

ordinal numbers set of scale of 100, the rank indicates the percentage of individuals in a group who fall at or below a given level of performance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

equal-unit scales

A

the difference between any two consecutive numbers reflect an equal empirical or demonstrable difference

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

additivity

A
  • within ratio scales, numbers achieve the property of additivity
  • thay can be added and the results are expressed as ratios and are meaningful
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Why is the meaning of numbers relevant to psychological testing?

A
  • Results of (most) psychological tests are expressed as scores, which are numbers that have specific meanings
  • Scale level of test scores is important because of (different) limitations in interpretability
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

statistics

A

branch of methematics dedicted to organizing, depicting, summarizing, analyzing and otherwise dealing with numerical data
- measurement derived from sample data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

descriptive statistics

A

Numbers and graphs used to describe, condense or represent data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

inferential statistics

A

used to estimate population values based on sample values or to test hypotheses

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

parameters

A

data derived from populations

- mathematically exact numbers

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Frequency distributions

A

organize raw data in a way they can be inspected (grouped frequency distributions - scores are grouped into intervals)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

cumulative percent column

A

consecutive additions of the numbers in the Percent column from lowest to thighest score

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

graphs

A

once transformed into a frequency distribution, they can be transposed into graphic format

26
Q

Measures of central tendency

A
  • mean
  • mode
  • median
27
Q

mode

A

mots frequent occuring value in a distribution, useful primarily when dealing with qualitative/categorial variables

28
Q

Median

A

value that divides a distributions that had been arranged in order of magnitude into twó halves

29
Q

Mean

A

obtained by summing all the values in a distribution and dividing the total by the number of cases in the distribution

30
Q

Measures of variability

A

range
semi-interquartile range
veriance and standard deviation

31
Q

Range

A

distance between two extreme point (highes and lowest value)

32
Q

semi-interquartile range

A

half of the interquartile range (IQR), which is the distance between the points that demarce the tops of the first anf third quarters of a distribution

33
Q

Variance

A

sum of the squared differences or deviations between each value (X) in a distribution and the mean of that distribution (M), divided by n

34
Q

Standard deviation

A

square root of variance

35
Q

the importance of variability

A

Greater variability leads to (potentially) more pronounced distinctions in the assessment of individuals‘ characteristics
- Psychologicals test need to be constructed with the goal of maximizing score differences among testees from a distinctly specified target population

36
Q

The normal curve model

A

Parametric description of a unimodal distribution using its mean and standard deviation

37
Q

properties of the normal curve model

A
  • bell-shaped
  • bilaterally symmetrical
  • unimodal
  • mean, median and mode that coincide at the center of the distribution
38
Q

Uses of Normal Curve Model

A
  1. descriptive uses

2. Inferential uses

39
Q

Descriptive uses (normal curve model)

A
  • to describe the characteristics of a score distribution

- to normalize test scores

40
Q

Inferential uses (normal curve model)

A
  • to estimate population parameters

- to test hypotheses about differences

41
Q

Normalizing scores

A

involves transforming them so that they have the same meaning, in terms of their positions, as if they came from a normal distribution

42
Q

Sampling distributions

A

are hypothetical and not real distributions of values assuming that an infinite number of samples of a given size could be drawn from a population

43
Q

Standard error (ER)

A

conceived of as the standard deviation of the sampling distribution that would result if we obtain the same statistic from a large number of randomly drawn samples of equal size

44
Q

Kurtosis

A

Degree of flat- or peakedness of a distribution

45
Q

Platykurtic

A

greatest amount of dispersion, manifested in tails that are more extended

46
Q

leptokurtic

A

least dispersion

47
Q

mesokurtiv

A

the normal distribution

48
Q

Skewness

A

Lack of symmetry

49
Q

negatively skewed

A

most of the values are at the top end of the scale and the longer tail extends toward the top of the scale (SK <0)

50
Q

positively skewed

A

most of the values are at the bottom and the longer tail extends toward the top of the scale (SK >0)

51
Q

Linear correlation

A

a correlation between variables is linear when the direction and the rate of Chang in one variable are constant with regard to the chnages in other variable

52
Q

coefficient of determination

A

proportion of variance shared by two variables is often estimated by squaring the correlation coefficient

53
Q

Correlation Coefficients

A
  • r, Spearman‘s rho
  • Phi
  • Point biserial r
54
Q

Necessary conditions for the use of r

A
  1. Pairs of observations are independent of each other
  2. Continous variables which are measured on interval or ratio scales
  3. (True) linear relationship between the variables
55
Q

correlation

A

the extent to which variables are related

56
Q

correlation coefficient

A

measures degree ad direction of the correlation, number that fluctuates between -1 and +1

57
Q

Heteroscedasticity

A

the dispersion or variablity in the scatterplots is not uniform throughout the range of values of two variables

58
Q

Spearman’s rho (r)

A

for when the variables to be correlated are in ordinal form

59
Q

phi

A

both variables are dichotomous

60
Q

point biserual r

A

when one of the variables to be correlated is dichotomous

Test Theory (5 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions