Exploring and Describing Survey Data: Classification, Frequency Distributions, and Data Visualization

Survey Data Analysis in Statistics

Introduction to Survey Data

Survey data is commonly used in statistics to gather information about a population's characteristics, preferences, or behaviors. In this example, a survey was distributed to prospective students to collect demographic and academic information, as well as their likelihood of attending a university. The analysis of such data involves classifying variables, organizing data into tables, and visualizing results for interpretation.

Classifying Variables

Types of Variables

• Qualitative (Categorical) Variables: Variables that describe qualities or categories. Examples: Housing (On campus/Off campus), Academic Program.

• Quantitative Variables: Variables that represent numerical values. Examples: Age, Likelihood to attend UMGC.

Levels of Measurement

• Nominal: Categories with no inherent order (e.g., Academic Program, Housing).

• Ordinal: Categories with a meaningful order but not evenly spaced (e.g., Likelihood to attend, if treated as ranks).

• Interval: Numerical values with meaningful differences but no true zero (e.g., Temperature in Celsius; not present in this survey).

• Ratio: Numerical values with meaningful differences and a true zero (e.g., Age).

Organizing Data: Frequency Distributions

Tabular Data from the Survey

The survey data can be organized into a frequency distribution table to summarize the responses for a particular variable. For example, the frequency of each housing type:

Housing	Frequency
On campus	7
Off campus	7

Additional info: This table helps quickly compare the number of students preferring each housing type.

Relative Frequency Table

Housing	Frequency	Relative Frequency
On campus	7	0.5
Off campus	7	0.5

Relative frequency is calculated as: $\text{Relative Frequency} = \frac{\text{Frequency}}{\text{Total Number of Observations}}$

Data Visualization

Choosing a Graph Type

• Bar Graph: Useful for displaying frequencies of categorical variables (e.g., Housing type, Academic Program).

• Pie Chart: Shows proportions of categories as parts of a whole.

• Histogram: Best for quantitative, continuous data (e.g., Age).

• Dot Plot/Box Plot: Useful for visualizing distributions and identifying outliers in quantitative data.

Example: A bar graph is an effective choice for visualizing the frequency of students in each academic program, as it clearly shows the number of students per category.

Measures of Central Tendency and Dispersion

Definitions

• Mean: The arithmetic average of a set of values. $\text{Mean} = \frac{\sum x_i}{n}$

• Median: The middle value when data are ordered from least to greatest.

• Mode: The value that appears most frequently in the data set.

Measures of Dispersion

• Range: The difference between the maximum and minimum values. $\text{Range} = \text{Max} - \text{Min}$

• Standard Deviation: A measure of how spread out the values are from the mean. $s = \sqrt{\frac{\sum (x_i - \bar{x})^2}{n-1}}$

Example: For the variable "Age," you can calculate the mean, median, and mode to summarize the central tendency, and the range and standard deviation to describe the spread of ages among respondents.

Interpretation of Dispersion

• A narrower standard deviation indicates that the data points are closer to the mean, suggesting less variability.

• A wider standard deviation indicates more variability in the data.

• In the context of student ages, a narrow standard deviation would mean most students are of similar age, while a wide standard deviation would indicate a more diverse age range.

Summary Table: Survey Variables Classification

Variable	Type	Level of Measurement
Age	Quantitative	Ratio
Housing	Qualitative	Nominal
Academic Program	Qualitative	Nominal
Likelihood to attend UMGC	Quantitative (Ordinal if treated as ranks)	Ordinal

Conclusion

Organizing and analyzing survey data is a foundational skill in statistics. By classifying variables, constructing frequency tables, visualizing data, and calculating measures of central tendency and dispersion, you can extract meaningful insights from raw data and effectively communicate your findings.