- Practice
- About 17 min
- 5 practice questions
Types of statistics, kinds of data and the arithmetic mean
Distinguish descriptive from inferential statistics, classify data as quantitative or qualitative (nominal, rank) and primary or secondary, and calculate the arithmetic mean by the direct and short-cut methods.
If something looks wrong, please tell us.
By the end of this lesson you can
- State three differences between descriptive and inferential statistics.
- Classify data by characteristic (quantitative or qualitative) and by source (primary or secondary).
- Find the arithmetic mean of raw data by the direct method and by the short-cut (assumed mean) method.
Before we calculate anything we must know what kind of numbers we have and what we want to do with them: describe them, or use a sample to say something about a bigger group. This lesson sorts out the types of statistics and data and then calculates the most familiar summary figure, the mean.
Descriptive and inferential statistics
Two types of statistics
-
Descriptive statistics
- Summarises and describes data
- Describes the group under study
- Uses charts, diagrams and graphs
- Gives a summary of the data
-
Inferential statistics
- Uses sample data to infer about a population
- Draws conclusions beyond the sample
- Uses hypotheses, testing and prediction
- Tries to understand the population
Kinds of data
Data are facts or numbers collected to be examined and used for decisions; they are the raw material of statistics. The book classifies them in two ways.
Classification of data
-
By characteristic
- Quantitative: measured in definite units (age, income, number of firms)
- Qualitative: presence or absence of an attribute (gender, community, honesty)
- Qualitative splits into nominal and rank data
-
By source
- Primary: collected for the first time from the original source
- Secondary: already exists, published or unpublished (for example data of CSO, NSSO, RBI)
Nominal data come from sorting units into categories by a quality: students as male or female, workers as skilled, semi-skilled or unskilled, employees by education level. Rank data come from giving ranks 1, 2, 3, ..., n by performance, as in a test, contest or interview. Data collected through your own questionnaire are primary; figures taken from a published report are secondary.
Averages and the arithmetic mean
A central value is called a measure of central tendency, an average or a measure of location: a typical value around which other figures cluster and which stands for the whole group. The book names five averages: mean, median and mode are simple averages; geometric mean and harmonic mean are special averages. This chapter calculates the arithmetic mean (X̄) for raw (ungrouped) data.
Direct method: X̄ = ΣX ÷ n. Short-cut method: X̄ = A + Σd ÷ n, where A is an assumed mean and d = X − A.
Worked example 1: direct method
Made-up data: daily wages (₹) of six workers are 420, 380, 450, 400, 510 and 480. Step 1: ΣX = 420 + 380 + 450 + 400 + 510 + 480 = 2,640. Step 2: n = 6. Step 3: X̄ = 2,640 ÷ 6 = ₹440. Check: 440 × 6 = 2,640.
Worked example 2: short-cut method
Made-up data: a student's marks in five subjects are 64, 71, 58, 80 and 77. Take the assumed mean A = 71.
| X | d = X − 71 |
|---|---|
| 64 | −7 |
| 71 | 0 |
| 58 | −13 |
| 80 | 9 |
| 77 | 6 |
| Total | Σd = −5 |
Step 1: Σd = −7 + 0 − 13 + 9 + 6 = −5. Step 2: X̄ = A + Σd ÷ n = 71 + (−5 ÷ 5) = 71 − 1 = 70. Check by the direct method: ΣX = 64 + 71 + 58 + 80 + 77 = 350, and 350 ÷ 5 = 70. Both methods agree, as they always must: the short cut only makes the numbers smaller.
Key takeaways: descriptive statistics summarises data; inferential statistics uses a sample to judge a population. Data are quantitative or qualitative (nominal or rank) by characteristic, and primary or secondary by source. Of the five averages, mean, median and mode are simple and GM and HM special. The mean is ΣX ÷ n, or A + Σd ÷ n with an assumed mean.
Check your understanding
-
1. Data collected by a researcher through her own questionnaire are:
- secondary data
- published data
- primary data
- grouped data
Show answer
Answer: C. Data collected for the first time from the original source are primary. Secondary data already exist in published or unpublished form.
-
2. Classifying workers as skilled, semi-skilled and unskilled gives:
- nominal data
- rank data
- quantitative data
- secondary data
Show answer
Answer: A. Sorting units into categories by a quality, with no order of size measured, gives nominal data. Rank data assign positions 1, 2, 3 by performance.
-
3. Which of these is a special average, not a simple average?
- Mean
- Median
- Mode
- Harmonic mean
Show answer
Answer: D. The book calls mean, median and mode simple averages, and geometric mean and harmonic mean special averages.
-
4. Made-up data: find the mean of 3, 7, 8, 12 and 15.
- 8
- 9
- 10
- 45
Show answer
Answer: B. ΣX = 3 + 7 + 8 + 12 + 15 = 45; n = 5; X̄ = 45 ÷ 5 = 9. 45 is the total, not the mean.
-
5. Made-up data: using the assumed mean A = 50 for the values 52, 48, 55 and 45, the mean is:
- 52.5
- 50
- 48.5
- 51
Show answer
Answer: B. d = 2, −2, 5, −5, so Σd = 0. X̄ = 50 + 0 ÷ 4 = 50. Check: (52 + 48 + 55 + 45) ÷ 4 = 200 ÷ 4 = 50.
Sign in to keep track of the lessons you finish.