Descriptive statistics in survey analysis

Analyzing Survey Data in Python

EbunOluwa Andrew

Data Scientist

What are descriptive statistics in survey analysis?

Basic measures used to describe survey data.
Consist of descriptions of single variables and associated survey sample.

magnifying glass and pen over graph

Why use descriptive statistics?

Allow data to summarized clearly
Forms
- Tabular presentations
- Visualizations
Help identify outliers

Graphs with magnifying

Frequency and distributions

Grouped data based on number of occurrences in each class
On qualitative and quantitative data
Count of different outcomes in a raw survey dataset
Bar charts, histograms, pie charts, line charts, etc

Looking at the statistical data, graphs, and charts

Central tendency: mean, median, mode

Single value reflecting center of data distribution
Mean = average value
Median = middle score for dataset in ascending order
Mode = most frequent value in dataset

Measures of variability

Determines how far apart data points appear to fall from center
Range
- Distance between highest and lowest values
Standard deviation
- Average variance
- Insight into distance between value in a dataset and mean value

Crowd from above forming a growth graph with lines connecting between them

Survey: dietary_habits

dietary_habits.head()

| Age   | Gender | meals_per_day | eat_out_per_wk |
|-------|--------|---------------|----------------|
| 18-24 | Male   |             5 |              4 |
| 18-24 | Male   |             4 |              1 |
| 45-54 | Male   |             3 |              3 |
| 18-24 | Male   |             2 |              1 |
| 18-24 | Female |             3 |              1 |

Frequency distribution: dietary_habits

dietary_habits.Gender.value_counts().to_frame("Number")

|        | Number |
|--------|--------|
| Male   | 40     |
| Female | 38     |

Index: Gender

Frequency distribution: dietary_habits

dietary_habits.Gender.value_counts().to_frame("Number").plot(kind='bar')

gender frequency distribution bar plot

Measures of central tendency: dietary_habits

.mean()
.median()
.mode()

Measures of central tendency: dietary_habits

.mean()

dietary_habits.mean()

| meals_per_day  | 3.128205 |
| eat_out_per_wk | 1.897436 |
| dtype: float64 |          |

Measures of central tendency: dietary_habits

.median()

dietary_habits.median()

| meals_per_day  | 3.0 |
| eat_out_per_wk | 1.5 |
| dtype: float64 |     |

Measures of central tendency: dietary_habits

.mode()

dietary_habits.mode()

| Age   | Gender | meals_per_day | eat_out_per_wk |
|-------|--------|---------------|----------------|
| 18-24 | Male   |             3 |              1 |

Measures of variability: dietary_habits

print(dietary_habits.meals_per_day.max() - dietary_habits.meals_per_day.min())

print(dietary_habits.meals_per_day.std())

0.6518500018473766

Let's practice!

Analyzing Survey Data in Python