Question

wledge check 1- frequency distributions + levels of measurement + graphi
frequency of sex during last year
...
cumulative
frequency percent valid percent percent
valid not at all 204 13.6 23.6 23.6
once or twice 90 6.0 10.4 34.0
once a month 105 7.0 12.2 46.2
2-3 times a month 141 9.4 16.3 62.5
weekly 134 8.9 15.5 78.0
2-3 per week 149 9.9 17.2 95.3
4+ per week 41 2.7 4.7 100.0
total 864 57.6 100.0
missing iap 587 39.1
dk 8 .5
na 41 2.7
total 636 42.4
total 1500 100.0

• what two(2) observations do you glean from this frequency distribution examining th

frequency of sex during the last year?

• what is the level of measurement for this variable?
• what is the best way to graphically display this data? explain.

he following frequency distribution, based on the 2014 general social survey (gss), examine\ttitudes about sex before marriage. use this distribution to answer the questions that follow
sex before marriage
...

Explanation:

First Question: Two Observations from the Frequency Distribution

1. Missing Data Significance: A large portion (42.4% total, 39.1% “IAP”) of respondents have missing data (e.g., “IAP,” “DK,” “NA”), indicating uncertainty or non - response about reporting sex frequency.
2. Most Common Category: Among valid responses, “NOT AT ALL” (23.6% valid percent) and “2 - 3 PER WEEK” (17.2% valid percent) are notable, but the highest valid frequency is “2 - 3 TIMES A MONTH” (141) and “2 - 3 PER WEEK” (149), while “NOT AT ALL” has the highest valid percent initially. Also, the cumulative percent shows that about 34% of valid respondents had sex ≤ once or twice, and 100% of valid respondents are accounted for in the 4+ PER WEEK category.

The variable “frequency of sex during last year” has ordered categories (NOT AT ALL < ONCE OR TWICE < ONCE A MONTH < … < 4+ PER WEEK) with meaningful rank order, and differences between categories (e.g., “ONCE A MONTH” vs “2 - 3 TIMES A MONTH”) are interpretable in terms of frequency. However, the intervals between categories (e.g., the difference between “ONCE A MONTH” and “2 - 3 TIMES A MONTH” in actual frequency) are not necessarily equal. This fits the ordinal level of measurement, where data is categorical, ordered, but with unequal or unquantified intervals between categories.

A bar graph (or a Pareto chart) is best. Ordinal data with distinct categories (the frequency levels) is well - represented by bar graphs, where each bar’s height corresponds to frequency/percent. A Pareto chart (sorted by frequency) can also highlight the most common categories. A bar graph is suitable because: 1) It preserves the ordinal order of categories. 2) It clearly shows the frequency/percent of each category, including the missing data categories if included. A histogram is not ideal here because the categories are not continuous (they are distinct, ordered groups, not a continuous scale with equal intervals).

Answer:

1. A substantial proportion (42.4% of total respondents) have missing data (e.g., “IAP,” “DK,” “NA”) regarding their sex frequency.
2. Among valid responses, “NOT AT ALL” has the highest initial valid percent (23.6%), and the “2 - 3 PER WEEK” category has a relatively high frequency (149); also, the cumulative percent shows that 34.0% of valid respondents reported having sex once or twice or not at all in the last year.

Second Question: Level of Measurement