First Steps: Organizing Data Looking at a set of raw data is like emptying the box of a jigsaw puzzle on a table… it’s just a meaningless jumble of pieces.
Organizing Data Frequency distribution - a table which shows how often each value of a variable occur. Consider the following data:
Ungrouped Frequency Distribution Score f n = 40
Grouped Frequency Distribution Score f n = 40 (interval width = 3)
Organizing Data Exact limits - the limits of a score are considered to extend one-half the distance between the smallest unit of measurement above and below the score Exact Limits
Grouped vs. Ungrouped - ungrouped distributions work best with small ranges - ungrouped distributions contain exact information for all values, but are more difficult to interpret. - grouped distributions usually provide a better “picture” of data, but lose some detail. - grouped distributions work best with large ranges
Other Frequency Distributions Score f % f ÷ 40 = ÷ 40 = n = 40 Relative frequency distribution - table that contains the proportion or percent for each value or interval. f/n = prop. f/n x 100 = %
Other Frequency Distributions Score f cum f n = 40 Cumulative frequency distribution - table that shows the accumulated frequencies below the upper exact limit of each interval
Other Frequency Distributions Score f cum f % cum f ÷ 40 = ÷ 40 = n = 40 Cumulative percent frequency distribution - table that shows the percent of accumulated frequencies below the upper exact limit of each interval. cum f/n = cum prop. f cum f/n x 100 = cum % f
Organizing Data Frequency distributions can also be constructed for qualitative data: Major f Biology 529 Chemistry 221 Physics 106 Geology 58 n = 914
Construct a relative frequency distribution of the following: Is a Stolen Kiss a Form of Sexual Harassment? Females Males Score f f n = 87 n = 59
Is a Stolen Kiss a Form of Sexual Harassment? Score % females % males n = 87 n = 59