Question

-1 206 13.7 20.6 27.5
2 261 17.4 26.1 53.5
3 173 11.5 17.3 70.8
4 131 8.7 13.1 83.9
5 53 3.5 5.3 89.2
6 49 3.3 4.9 94.1
7 9 .6 .9 95.0
8 22 1.5 2.2 97.2
9 1 .1 .1 97.3
10 8 .5 .8 98.1
12 10 .7 1.0 99.1
14 1 .1 .1 99.2
15 1 .1 .1 99.3
16 2 .1 .2 99.5
20 3 .2 .3 99.8
24 2 .1 .2 100.0
total 1001 66.7 100.0
missing jap 495 33.0
dk 4 .3
total 499 33.3
total 1500 100.0

• what two(2) observations do you glean from this frequency distribution examining the number of hours per day watching tv?
• what is the level of measurement for this variable?
• what is the best way to graphically display this data? explain.

Explanation:

First Sub - Question: Two Observations from the Frequency Distribution

1. Central Tendency and Distribution Shape: A large proportion of respondents (66.7% of the non - missing data, with 1001 out of 1500 non - missing) report relatively low numbers of hours watching TV per day. For example, the cumulative percentage reaches 95.0% when the number of hours is 7 or less, indicating that most people watch TV for a small number of hours daily.
2. Missing Data: There is a significant amount of missing data. The total number of missing cases (including IAP and DK) is 499 out of 1500, which is 33.3% of the total sample. This could potentially impact the generalizability of the results regarding TV - watching hours.

The variable "number of hours per day watching TV" is a ratio level of measurement. Ratio data has a true zero point (0 hours of TV watching means no TV watching at all), and we can perform operations like multiplication and division (e.g., we can say that someone who watches 4 hours a day watches twice as long as someone who watches 2 hours a day). Also, the data has equal intervals (the difference between 1 and 2 hours is the same as between 2 and 3 hours in terms of the concept of time measured).

A histogram is the best way to display this data. The variable "hours per day watching TV" is a quantitative (ratio - level) variable with discrete or potentially continuous values (since hours can be measured in fractions, although in the frequency distribution here they seem to be reported as whole numbers or small decimals). A histogram is suitable for quantitative data as it shows the frequency distribution of a continuous (or discrete - treated - as - continuous) variable by dividing the data into intervals (bins) and representing the frequency of each interval with a bar. This allows us to visualize the shape of the distribution (e.g., whether it is skewed, has peaks, etc.), which is important for understanding the pattern of TV - watching hours. A bar graph is not as appropriate here because bar graphs are better for categorical variables, while a histogram is designed for quantitative data with intervals.

Answer:

1. Most respondents (66.7% of non - missing) watch TV for a small number of hours daily (e.g., 7 or fewer hours account for 95.0% of non - missing cumulative percentage).
2. There is a significant amount of missing data (33.3% of the total sample, 499 cases), which may affect result generalizability.

Second Sub - Question: Level of Measurement for the Variable (Hours per Day Watching TV)