Chapter 2 Flashcards
Inferences:
Logical deductions about events that can’t be observed directly.
Descriptive Statistics
Methods used to provide a concise description of a collection of quantitative information.
Inferential Statistics:
Methods used to make inferences from a small group of people, known as a sample, to a large group of people, known as a population.
Psychological Testing
based on Inferential Statistics.
Measurement:
The application of rules for assigning numbers to objects.
-Magnitude: (scale)
-Magnitude: “moreness” (such as “taller,” “shorter”
-Equal Intervals:(scale)
: The difference between two points next to each other is the same (such as inches)
-Absolute 0: (scale)
It is possible to have none of a particular quality being measured (a heart rate of 0 means the patient is dead)
*Nominal Scales (type of scale)
Not really scales, just assigning random designator numbers to people or things (football jerseys)
-no properties of of a numbered scale
*Ordinal Scales:(type of scale)
Have the property of “moreness” but not equal intervals or absolute 0 (like “taller”)
*Interval Scales:(type of scale)
Have magnitude and equal intervals, but not necessarily absolute 0–temperature in ºF or ºC, for example
*Ratio Scales: (type of scale)
Have magnitude, equal intervals and absolute 0
*frequency distribution
summarizes the scores of a group of individuals in terms of how frequently each value was obtained (horizontal axis from lowest to highest)
bell-shaped curve
with a mean where the median and the mode are. If it’s skewed, the median isn’t at the mode
positive skew
means more scores are above the mode than below it
negative skew
means more scores are below the mode than above it
class interval
the distance between two consecutive measurements. This is a value that you decide is the meaningful difference between one measurement and the next higher one.
Examples: Temperature, money, length (metric or British)
Percentile Ranks:
A percentile rank answers the question “what percent of the scores fall below a particular score (X1)?”
Importance of percentile rank
because that is the information that an IQ score gives you—that is, how your intelligence score compares to that of other people; where you are in the total distribution of IQ scores. Deviation IQ is based on this
Percentiles
are the specific scores or points within a distribution. They divide the total frequency for a set of observations into hundredths
Difference between percentile and percentile rank
Percentile is a measure of relative performance–you must specify the population to which the score in question is being compared
*Mean:
The average score in a distribution
The mean = the sum of all scores divided by the number of cases.
Median:
The point where half the cases are above it and half below
Mode:
The “bump,” the point where the largest number of cases fall
*Standard Deviation:
An approximation of the average deviation around the mean.
*Variance:
The variation of the scores around the mean; calculated as the squared deviation around the mean.
Variance
σ2 = Σ(X – X)2
N
(Variance is easier to calculate, but SD (or σ) is more meaningful, more important)
σ
refers to the standard deviation of a population
S
refers to the standard deviation of a sample drawn from that population.
sample from population
A subset of the population, drawn to represent it in an unbiased (random) fashion
Z-scores
indicate a score’s percentile rank in terms of the standard deviation of the distribution
Z =
distance of a score from the mean in standard deviations, with scores above the mean stated as positive Z-scores and scores below the mean stated as negative Z-scores
A Z score of 1.0 means the person’s score
approximately the 84th percentile.
A Z score of -1.0 is at
approximately the 16th percentile.
Wechsler IQ Tests
the most commonly used individual IQ tests, have been defined to have a Standard Deviation of 15 because that was closest to the Standard Deviation of the scores on the Stanford-Binet, calculated by the formula of Mental Age ÷ Chronological Age X 100.
deviation IQ score
The score tells you where you are in the distribution of other individuals your age
standard normal distribution
(the “bell-shaped curve”) is also called the symmetrical binomial probability distribution
Units on the X (horizontal) axis are
usually given in Z units
McCall’s T
In 1939 W.A. McCall developed a system in which scores on standardized exams were presented as standard deviations, where instead of a standard deviation being equal to 1, as it is with Z-scores, it’s equal to 10.
He set the mean at 50 rather than at zero, so that one standard deviation below the mean is 40, and one standard deviation above the mean is 60.
who uses mccall’s T system
MMPI scaled scores are still graphed using this system. SAT and GRE scores are the same except that μ = 500 and Z=100.
how can you get tscores
applying the formula T = 10Z + 50 (and you can calculate back the other way if you want to).
Quartile systems
divided the percentage scale into four groups (“You’re in the top quarter of…”)
decile systems
divide the scale into tenths.
“interquartile range”
is the middle 50% of scores
Stanines divide the scores into “standard 9ths”
*Norm-referenced tests
compare individuals to a normative group (and therefore if you can make somebody else look worse, you look better).
*Criterion-referenced tests
compare an individual to a criterion on specific types of skills, tasks, or knowledge (so making somebody else look bad doesn’t help you.)