Intro to Research Ch. 8

ordinal level of measurement

(rank ordered) ranked high to low or use of a rating scale. often uses numbers but can sometimes use traits

3 characteristics of ratio level measurement:

1. ability to arrange numbers in a continuum 2. ability to specify amount and differences between amounts 3. ability to identify an absolute zero relative to a characteristic

descriptive statistics

a way of summarizing data to convey the main info

frequency polygon

a way to examine the shape of your distribution

explain when a researcher would use each of the following: a. scatterplot b. column graph c. line graph

a. scatterplot- relationship between 2-3 continuous measures b. column graph- comparing magnitude of one or more variables c. line graph- depicting nonlinear relationships (interactions)

difference

actual range- highest score-lowest score

nominal level of measurement

aka naming level. researchers assign participants and their responses to a category

pie chart

appropriate for nominal, categorical measure. useful for percentages or proportions

mean

average

what descriptive statistics are used most often with nominal level measures?

categories

match the level of measurement with its brief description: ___nominal ___ordinal ___interval ___ratio a. rank order from highest to lowest b. shows how much particpants differ but no true 0 c. measures with a true 0 d. naming or mutually exclusive categories

d. nominal a. ordinal b. interval c. ratio

standard deviation

dispertion of scores around the mean. (find the mean then the amount between each score and the mean)

positive skew

distribution has an abnormally long tail that stretches in the positive direction (in regard to outliers)

negative skew

distribution that has an abnormally long tail that stretches in the negative direction (i regard to outliers)

variance

divide the sum of squares by the number of participants (standard deviation is the square root of variance)

proportion

dividing # for particular category b the total #, yielding a number between 0.0 and 1.0

percentage (%)

dividing the # participants/observations by total # of particpants then x 100

mutually exclusive category

each participant or instance of a behavior fits into only one category

exhaustive category

if every participant or every behavior under observation fits into a category

XY Scatterplot

illustrates the relationship between two sets of scores usually from the same participant

column and bar graphs

illustrating th magnitude or frequency of one or more variables

the scores obtained from articulation/phonology or language test most likely measure at what level?

interval

subjects of a study indicate where they live, which researcher classified by geographic regions. what scale of measurement would this be?

nominal

median

number that occurs at the midpoint

researchers classified preschool children's history of middle ear infections as frequent/moderate/infrequent based on the number of reported episodes. what scale of measurement would this be?

ordinal

what kind of chart or graph could you use if your level of measurement was nominal and you had percentages that totaled to 100%?

pie chart

sampling error

possibility that a random sample might yield an estimate of a population characteristic that is not perfectly accurate

interval level of measurement

provides info on which participants have higher or lower scores and how much they differ

what measure of validity is calculated by determining the difference between the minimum and maximum scores?

range

will an extreme, outlying score have a greater impact on the range or the standard deviation of a set of scores?

range

what level of measurement would allow you to say that a score of 30 was three times as much as a score of 10

ratio

margin of error

reflects the potential distribution of means

range

reporting the minimum and maximum scores

interquartile range

represents the difference of the scores at the 75th and 25th percentiles

ratio level of measurement

same as interval, only with a true zero. (often assessments of physical attributes)

mode

score that occurs most often

bimodal

scores that occur most often in a set

sum of squares

squaring the differences (standard deviation) to eliminate the negative numbers

normal distribution

the dispersion of scores around the mean is symmetrical and half of the scores fall above the mean and half fall below

sampling distribution

the potential distribution of means

outliers

the presence of scores that separate quite a bit form the other scores

what is the relationship between standard deviation and variance

the standard deviation is the square root of the variance

line graph

useful for illustrating different values for frequencies, counts, percentages, and averages

scatterplot

useful for illustrating the relationship between two and even three continuous measures.