Probability and Statistics

The relationship between probability and statistics

It is important to understand the differences and similarities between probability and statistics. They are different but related subjects.

Probability is a theoretical subject used to analyse the likelihood of events happening in the future. On the other hand, statistics is an applied subject which uses probability theory to analyse data which has been collected.

Probability and statistics are related in that the theories we develop in probability mathematics are compared with statistical findings which can tell us more information about the data. We can also use statistics to estimate the probability of something happening in the future.

Probability

Probability is the likelihood of some event occurring. The probability of an event happening is always a fraction, decimal or percentage between 0 and 1.

The basic formula for probability is the following:

Some terminology for probability

There is a lot of terminology in probability which you need to be familiar with. The better you understand the basic concepts detailed below, the easier it will be when you tackle trickier probability questions!

Terminology	Definition
Experiment or trial	A repeatable action which has a defined set of outcomes.
Sample space	The set of all possible outcomes of an experiment, usually denoted by or S.
Event	An event is one particular outcome, usually denoted by a capital letter. The probability of event A occurring is commonly notated as P(A).

An example of probability

You are on a game show and you have to choose between 2 boxes. One contains a sports car and the other is empty. You are given no clues as to which box contains the prize and must therefore choose a box at random. What is the probability you win the sports car?

Solution

Here, the sample space is S = {W, L} where W stands for 'win' and L stands for 'lose'. This is the sample space since these are the only possible outcomes.

Intuitively, since you are choosing at random, there is an equal probability of choosing the prize and choosing the empty box. Since probabilities are always between 0 and 1, the probability of choosing the sports car is 0.5 (= 50%).

We can also arrive at this answer by using the basic formula for probability. The number of ways to choose the box with the sports car is 1, and the total number of outcomes (either choosing the sports car or choosing the empty box) is 2. Using mathematical notation, the probability of winning is as follows:

.

Therefore, the probability of choosing the sports car and therefore winning is which is equivalent to 50% or 0.5.

Statistics

Data usually contains far too many individual numbers for the human brain to even comprehend, let alone understand. This is why we need statistics: to help us to better understand the underlying complexities in the data we have collected.

The sorts of tasks you will have in statistics questions will fall into one of three categories: data collection, analysis, and visualisation and interpretation.

Data collection

The type of data collected will depend on the statistical question you want answered, i.e. the characteristic you want to study. You might be interested in collecting data through holding a scientific experiment (e.g. measuring the effect of location on the growth of plants) or perhaps by collecting data through observation (e.g. recording the number of students that are late for class).

All data falls into two categories: qualitative and quantitative.

Qualitative data refers to descriptive data, such as words. It might be collected through interviews or surveys.

Quantitative data refers to things which can be represented using numbers, like measurements such as height and weight, or quantities of something that we are interested in knowing more about. Important methods include:

• conducting a census

  • data is collected from every single member of a population

• sampling

  • data is collected from a subset of a population called a sample

  • this subset is either randomly chosen or is chosen to be representative of the wider population

• controlled experiments

  • a scientific procedure is planned such that accurate data can be collected and analysed

Quantitative data also falls into two categories: continuous and discrete.

Continuous variables can take on any value within a range, which can make it more difficult for us to count them. For example, length can be measured to as many decimal places as is possible. It's up to you, the data collector, to decide how many decimal places are necessary.

Discrete variables must take on particular values within a range, which makes it easier to count them. For example, shoe size can be 4 or 4.5, but not 4.26735!

Analysis

Once you have collected your data, it is now time to start analysing it. Since it is very difficult (and often impossible!) to understand data in its raw form, we need to condense it into manageable descriptions that retain as much information about the data as possible whilst being understandable. This is the ultimate goal of statistics: to understand, describe and find meaningful information from datasets.

Descriptive statistics are particularly useful. These are numbers that tell us something about the data. There are two kinds: measures of location and measures of spread.

Measures of location use one statistic to summarise a dataset. They include:

• mean

  • a particularly common and useful statistic that requires summing values and dividing by the number of values, often denoted by (pronounced 'x bar'):

;

• median

  • the middle value in an ordered list of values;

• mode

  • the most common value.

Measures of spread describe the variability of the data. They include:

• range

  • this is the largest value minus the smallest value in the data;

• interquartile range

  • this describes the range of the central 'quartiles' of the data, between 25% and 75% which surround the median at 50%.

Data is often be represented in frequency tables. Frequency refers to the number of times something occurs, which in the case of statistics will be the number of times a particular data value occurs. From a frequency table, you will be able to extract descriptive statistics such as the ones listed above. You will also be able to use these to visualise data.

Visualisation and interpretation

Descriptive statistics are useful in condensing data into a small amount of information, and can tell us about the location and spread of the data. Data visualisations, on the other hand, are able to graphically represent the data. For a thorough analysis, you would ideally use both methods to be able to fully understand the behaviour of the data.

We will now go through some examples of different data visualisations. Don't worry if you can't yet fully understand the examples below – there are in-depth explanations in other articles!

Line graphs

These are useful in representing continuous data and trends.

The following is an example of a line graph:

StudySmarter originals, Rebecca Farthing

Bar charts

These are useful in representing data which is grouped.

The following is an example of a bar chart:

StudySmarter originals, Rebecca Farthing

Histograms

These are useful in representing the frequency of something happening.

The following is an example of a histogram:

StudySmarter originals, Rebecca Farthing

Pie charts

These are useful in representing proportional data.

The following is an example of a pie chart:

StudySmarter originals, Rebecca Farthing

Box-and-whisker plots

These visualise the range, interquartile range and median.

The following is an example of a box-and-whisker plot:

StudySmarter originals, Rebecca Farthing

Scatter graphs

Shows the relationship between two variables.

The following is an example of a scatter plot:

StudySmarter originals, Rebecca Farthing

Finally, once the data has been analysed and represented graphically, we can draw conclusions from the data. Have a look at the example below.

An example of statistics

The following is data collected from a team's football matches. The frequency of the number of goals scored by each player per match is recorded from a total of 20 matches.

Team member	Scored 0 goals	Scored 1 goal	Scored 2 goals	Scored 3 goals	Scored 4 goals
Zack	5	6	4	4	1
Josh	9	5	7	0	0
Amy	0	13	3	4	0
Ahmed	2	11	4	1	2
Emily	7	12	0	0	1

a) What type of data is presented here?

b) Which team member scores the largest number of goals over the course of the season?

c) Which player scored the highest average number of goals per game?

d) Represent this data using a pie chart.

e) What does this data tell us about the performance of the players?

Solution

a) Frequency and goals scored per match are both quantitative, discrete data. It is impossible to score 0.5 goals!

b) By multiplying the number of goals by the frequency, f, we can find the total goals scored by each player:

Team member	f x 0	f x 1	f x 2	f x 3	f x 4	Total goals
Zack	0	5	4 × 2 = 8	4 × 3 = 12	2 × 4 = 8	33
Josh	0	5	7 × 2 = 14	0 × 3 = 0	0 × 4 = 0	19
Amy	0	13	3 × 2 = 6	4 × 3 = 12	0 × 4 = 0	32
Ahmed	0	11	4 × 2 = 8	1 × 3 = 3	2 × 4 = 8	30
Emily	0	12	0 × 2 = 0	0 × 3 = 0	1 × 4 = 4	16

We can see that the most goals scored by one player was by Zack, who scored 33 during the season.c) To find the average number of goals per match, we need to divide the total number of goals by the number of matches:Zack: goals per match

Josh: goals per match

Amy: to 2 d.p. goals per match

Ahmed: goals per match

Emily: goals per match

d) A pie chart shows proportional data. This means we need to work out what proportion of the total goals in the season are scored by each player. To do this, we should divide each player's total goals by the total number of goals scored during the season.Total number of goals scored by all players = 33 + 19 + 32 + 30 + 16 = 130Zack: Josh: Amy: Ahmed: Emily:

StudySmarter originals, Rebecca Farthing