Question

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

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

sex before marriage
...
valid | always wrong | 184 | 12.3 | 18.7 | 18.7

almst always wrg	60	4.0	6.1	24.8
sometimes wrong	168	11.2	17.1	42.0
not wrong at all	570	38.0	58.0	100.0
total	982	65.5	100.0

missing | iap | 495 | 33.0 | |

dk	19	1.3
na	4	.3
total	518	34.5

total | | 1500 | 100.0 | |

• what two(2) observations do you glean from this frequency distribution examining attitudes about sex before marriage?
• what is the level of measurement for this variable?
• what is the best way to graphically display this data? explain.

following frequency distribution, based on the 2014 general social survey (gss), examines...

Explanation:

1. Two Observations:

• The majority of valid respondents (58.0% with a frequency of 570) believe that sex before marriage is "NOT WRONG AT ALL".
• A significant portion of the total sample (34.5% with a frequency of 518) has missing data (including IAP, DK, NA), indicating that over a third of respondents either had an inappropriate response, didn't know, or the data was not applicable.

2. Level of Measurement:

The variable "attitudes about sex before marriage" has categories ("ALWAYS WRONG", "ALMST ALWAYS WRG", "SOMETIMES WRONG", "NOT WRONG AT ALL") that are ordered (from most negative to most positive attitude). So, the level of measurement is ordinal. This is because the categories have a meaningful order (ranking of attitudes) but the differences between the categories are not necessarily equal or measurable in a quantitative sense (e.g., we can't say how much more "wrong" "ALWAYS WRONG" is compared to "ALMST ALWAYS WRG" in a numerical way).

3. Best Graphical Display:

A bar graph (or a Pareto chart) would be the best way to display this data. Here's why:

• The variable is ordinal (categories with order) and we are dealing with frequencies/percentages of each category. A bar graph can clearly show the frequency (or percentage) of each attitude category.
• For ordinal data, a bar graph allows us to easily compare the heights of the bars for each category, which helps in visualizing the distribution of attitudes. If we use a Pareto chart, we can also order the bars from highest to lowest frequency (which is useful here as "NOT WRONG AT ALL" is the most frequent valid response) and include a cumulative frequency line, which aligns with the cumulative percent data provided. A pie chart, while it can show proportions, is less effective for ordinal data as it doesn't emphasize the order of categories as well as a bar graph, and it can be difficult to compare slice sizes visually. A histogram is not appropriate here because histograms are used for continuous or interval/ratio level data with grouped numerical values, while we have distinct categorical (ordinal) groups.

Answer:

s:

• Two Observations:

1. 58.0% (570) of valid respondents think sex before marriage is "NOT WRONG AT ALL".
2. 34.5% (518) of the total sample has missing data.

• Level of Measurement: Ordinal.
• Best Graphical Display: Bar graph (or Pareto chart); it shows frequency/percent of each ordinal category clearly for comparison.