Descriptive Statistics Exercises STUDYSmarter

Question 1

Small data sets don’t tell us very much and fake data sets tell us even less! However, it’s
useful to get some practice with the mean and standard deviation formulas to understand
how they work and what they measure.

For the following small, fake samples, do the following:

1. Draw a rough plot of the data values on a number line.

2. Calculate the mean, variance and standard deviation.

(a) 2, 5, 2, 3 (b) 2, 18, 1, 3, 1 (c) 2, 3, 1, 3, 1

[PSYC1101 students don’t have to work out means and variances by hand but it’s useful
to do it for a couple of small samples to become comfortable with the concept.]

Question 2

Now for some real data.

In 1898 twenty-four dead male house sparrows were brought to the Anatomical Laboratory
of Brown University after a severe winter storm. Their weights (in grams) were as follows
(in numerical order):

24.6, 24.6, 24.9, 25.0, 25.0, 25.1, 25.5, 25.6, 25.6, 25.8, 25.9, 26.0, 26.0, 26.0, 26.0, 26.1,
26.5, 26.5, 27.1, 27.5, 27.6, 28.3, 28.3, 31.1

As part of a study into why they died, the weights (in grams) of 35 house sparrows who
survived the storm were recorded (also listed in numerical order):

23.2, 23.6, 23.7, 23.8, 23.9, 24.1, 24.2, 24.3, 24.3, 24.5, 24.6, 24.7, 24.7, 24.8, 24.9, 25.4,
25.6, 25.7, 25.7, 25.7, 25.7, 25.9, 26.2, 26.2, 26.3, 26.3, 26.5, 26.6, 26.7, 26.7, 26.9, 26.9,
27.0, 27.9, 28.0

Think of each data set as a sample from a larger population of house sparrows.

Source: Ramsey, F., & Schafer, D. (1997) The Statistical Sleuth, ITP 1997, p.51

G Coates 1 July 2016

Descriptive Statistics Exercises STUDYSmarter

(a) From the plots, do you notice any differences between the distribution of weights of
the two groups of sparrows?

(b) Find the median for each sample.

(c) Find the interquartile range for each sample. [Not covered in PSYC1101.]

The means and standard deviations for both groups are given below (don’t try to work
them out yourself on paper!):

Dead sparrows: x = 26.3 g s = 1.47 g

Surviving sparrows: x = 25.5 g s = 1.26 g

(d) For each data set, work out the interval covered by “two standard deviations from
the mean” (in algebraic terms x ± 2s). What percentage of the values in each data
set is included in the interval?

Question 3 [STAT1520 only]

In this question we check whether the linear transformation formulas really work.

(a) Multiply each of the data values in Question 1(a) by −2 and then add 5.

(b) Work out the mean, variance and standard deviation of this new data set and
verify that they are related to the original mean and standard deviation in the ways
predicted by a linear transformation.

Question 4 [STAT1520 only]

A practical situation where a linear transformation might be useful.

To convert a temperature from degrees Celcius (C) to Farenheit (F ), we use the formula
9
F = C + 32
5
You have a large data set of temperature measurements from an experiment recorded in
degrees Celcius and have worked out that x = 28.6◦ C and s2 = 4.0(◦ C)2 .

If the data values were converted to degrees Farenheithat, what would their mean and
variance be?

G Coates 2 July 2016

Descriptive Statistics Exercises STUDYSmarter

Question 5

A class test has a mean of 70% and a standard deviation of 6%.

(a) Find the corresponding z−score for each of the following student’s marks and com-
ment on how well they performed relative to the class.

(i) 82 (ii) 55 (i) 70

(b) If you are told that a student has a z−score of 1.4, what was their original score?

Question 6

My daughter is nearly 11 years old and 165 cm tall. I am somewhat older and 187 cm
tall.

The mean height for girls her age is 144 cm with a standard deviation of 7.0 cm.

The mean height for men my age is 176.3 cm with a standard deviation of 6.8 cm.

Which of us is taller relative to our peers?

G Coates 3 July 2016

Descriptive Statistics Exercises STUDYSmarter

Solutions

Question 1

(a)

x = 3.0 units, s2 = 2.0 units, s = 1.414 units (to 2 d.p.)

(b)

x = 5.0 units, s2 = 53.5 units, s = 7.31 units (to 1 d.p.)

(c)

x = 2.0 units, s2 = 1.0 units, s = 1.0 units (to 1 d.p.)

Notice the effect one outlying value has on the mean and standard deviation.

Question 2

(a) In general, it seems the sparrows who survived tended to be lighter than those
who died (although a number of heavier sparrows survived and some lighter ones
perished). There is also an unusually heavy sparrow (outlier) amongst the dead
sparrows.

(b) Dead sparrows: median = 26.0 g

Surviving sparrows: median = 25.7 g

(c) Dead sparrows: IQR = 1.5 g

Surviving sparrows: IQR = 2.2 g

(d) Dead sparrows: 23.36 to 29.24 g. 100% (ie. all of them).

Surviving sparrows: 22.98 to 28.02 g. 95.8% (ie. 23 out of 24)

G Coates 4 July 2016

Descriptive Statistics Exercises STUDYSmarter

Question 3

(a) Data values become −4, −10, −4, −6

(b) x = −1.0 units, s2 = 8.0 units, s = 2.828 units (to 3 d.p.).

The linear transformation of the original xi values is −2xi + 5 so the mean of the
transformed data should be

−2 × 3.0 + 5 = −1.0 X.

The variance of the transformed data should be

(−2)2 × 2.0 = 8.0 X.

The standard deviation of the transformed data should be

| − 2| × 1.414 = 2.828 X.

Question 4

9
The mean would be × 28.6 + 32 = 83.48 ◦ F
5
 2
9
The variance would be × 4.0 = 12.96 (◦ F )2
5

Question 5

(a) (i) 2 (better than most of the class)

(ii) −2.5 (probably the worst in the class)

(iii) 0 (right on average)

(b) original score − 70

1.4 =
6
6 × 1.4 = original score − 70
6 × 1.4 + 70 = original score
78.4 = original score

G Coates 5 July 2016

Descriptive Statistics Exercises STUDYSmarter

Question 6

165 − 144
Daughter: z = = 3.0
7.0
187 − 176.3
Me: z = = 1.6
6.8

My daughter is taller relative to her peers. In fact she’s very tall indeed for her age!

Using STUDYSmarter Resources

This resource was developed for UWA students by the STUDYSmarter team for the
numeracy program. When using our resources, please retain them in their original form
with both the STUDYSmarter heading and the UWA crest.

G Coates 6 July 2016