Sampling and Basic Descriptive

Statistics. Basic concepts and

Techniques.

Lecture 6
Leah Wild
Overview

 Sampling In Quantitative Research

 Basic Descriptive Statistics And

Graphical Representation Of Data

 Quantification, Variables And

Levels Of Measurement
Sampling In Quantitative
Research

 Total Population
 Representative sample
 Probability Sampling
 Non-Probability Sampling
 Sample Size
Total Population

 The total collection of units, elements or

individuals that you want to analyse.
 These can be countries, lab-rats, light bulbs,
university students, banks, residents of a
particular area, regional health authorities
etc.
 The population for a study of infant health
might be all children born in the U.K. in the
1980's.
Sample
 A sample is a group of units selected from a larger
group (the population). By studying the sample it is
hoped to draw valid conclusions about the larger
group.
 Using example for study of infant health the sample
might be all babies born on 7th May in any of the
years.
 samples selected because the population is too
large to study in its entirety.
 Important that the researcher carefully and
completely defines the population, including a
description of the members to be included
Representative sample

 A sample whose characteristics correspond

to, or reflect, those of the original population
or reference population
 To ensure representativeness, the sample
may be either completely random or
stratified depending upon the
conceptualized population and the sampling
objective (i.e., upon the decision to be
made).
 A thorny issue in the social sciences- is it
possible to achieve?
Probability Sampling
A probability provides a quantitative description of the likely

occurrence of a particular event.

 A probability sampling method is

any method of sampling that uses
some form of random selection. In
order to have a random selection
method, you must set up some
process or procedure that assures
that the different units in your
population have equal probabilities
of being chosen (Clark 2002: 37).
Most Common Types of Probability
Sampling

 Simple Random Sampling

 Stratified Random Sampling
 Systematic Random Sampling
 Cluster Or Multistage Sampling
Simple Random Sampling

 where we select a group of subjects (a sample) for

study from a larger group (a population). Each
individual is chosen randomly and each member of
the population has an equal chance of being
included in the sample.
 Every possible sample of a given size has the same
chance of selection; that is, each member of the
population is equally likely to be chosen at any stage
in the sampling process. (Easton & Mc Coll 2004).
 A lottery draw is a good example of simple random
sampling. A sample of 6 numbers is randomly
generated from a population of 45, with each
number having an equal chance of being selected.
Stratified Random Sampling
 Often factors which divide up the population into sub-
populations (groups / strata)
 measurement of interest may vary among the different sub-
populations.
 This has to be accounted for when we select a sample from
the population to ensure our sample is representative of the
population.
 This is achieved by stratified sampling.
 A stratified sample is obtained by taking samples from each
stratum or sub-group of a population.
 Suppose a farmer wishes to work out the average milk yield
of each cow type in his herd which consists of Ayrshire,
Friesian, Galloway and Jersey cows. He could divide up his
herd into the four sub-groups and take samples from these
(Easton and Mc Coll 2004).
Systematic Random Sampling
 Systematic sampling, sometimes called interval sampling, means that
there is a gap, or interval, between each selection.
 Often used in industry, where an item is selected for testing from a
production line (say, every fifteen minutes) to ensure that machines and
equipment are working to specification.
 Alternatively, the manufacturer might decide to select every 20th item on
a production line to test for defects and quality. This technique requires
the first item to be selected at random as a starting point for testing and,
thereafter, every 20th item is chosen.
used when questioning people in surveys eg market researcher
selecting every 10th person who enters a particular store, after selecting
a person at random as a starting point;
 interviewing occupants of every 5th house in a street, after selecting a
house at random as a starting point.
If researcher wants to select a fixed size sample. In this case, it is first
necessary to know the whole population size from which the sample is
being selected. The appropriate sampling interval, I, is then calculated
by dividing population size, N, by required sample size, n, as follows:
 If a systematic sample of 500 students were to be carried out in a
university with an enrolled population of 10,000, the sampling interval
would be:
 I = N/n = 10,000/500 =20
Cluster Or Multistage Sampling

 Cluster sampling is a sampling technique where the

entire population is divided into groups, or clusters,
and a random sample of these clusters are
selected. All observations in the selected clusters
are included in the sample.
 every element should have a specified (equal)
chance of being selected into the final sample.
 typically used when the researcher cannot get a
complete list of the members of a population they
wish to study but can get a complete list of groups
or 'clusters' of the population
 Cheap, easy economical method of data collection.
Non-Probability Sampling

Main Types
 Convenience/ opportunity/accidental
sampling.
 Purposive/ judgemental sampling
 Quota sampling
 Snowball sampling
Convenience/ opportunity/accidental
sampling.

 volunteer samples
 Sometimes access through contacts or
gatekeepers
 ‘easy to reach’ population.
Purposive/ judgemental sampling

 Involves selecting a group of people

because they have particular traits that
the researcher wants to study
 e.g. consumers of a particular product
or service in some types of market
research
 My own questionnaire research on
‘New-Age’ Travellers.
Quota sampling

 widely used in opinion polls and market

research.
 Interviewers given a quota of subjects of
specified type to attempt to recruit.
 eg. an interviewer might be told to go out
and select 20 male smokers and 20 female
smokers so that they could interview them
about their health and smoking behaviours .
Snowball sampling
 Involves two main steps.
1. Identify a few key individuals
2. Ask these individuals to volunteer to
distribute the questionnaire to people
who know and fit the traits of the
desired sample (e.g. my research on
Travellers)
Sample Size
 In general, the larger the sample size (selected with
the use of probability techniques) the better. The
more heterogeneous a population is on a variety of
characteristics (e.g. race, age, sexual orientation,
religion) then a larger sample is needed to reflect
that diversity. (Papadopoulos 2003)
 Response rates vary on the type of surveys (e.g.
mail surveys, telephone surveys). Response rates
under 60 or 70 per cent may compromise the
integrity of the random sample. (ibid)
Basic Descriptive Statistics And
Graphical Representation Of Data

 Can be divided into two types

 Descriptive.
 Inferential
 Some authors suggest a third type
‘Associative’ (Downey 1975)
Descriptive Statistics
 Statistics which describe attributes of a
sample or population.
 includes measures of central tendency
statistics (e.g., mean, median, mode),
frequencies, percentages. minimum,
maximum, and range for a data set,
variance etc.
 organise and summarise a set of data
Inferential Statistics
 Used to make inferences or judgments about a larger
population based on the data collected from a small
sample drawn from the population.
 Eg Exit polling used during US elections to determine
how the population of voters voted
 A key component of inferential statistics is the calculation
of statistical significance of a research finding.
 used to determine whether changes in a dependent
variable are caused by an independent variable (Clark
2004)
 (HOMEWORK- WHAT ARE SOME OF THE
PROBLEMS ASSOCIATED WITH THESE KIND OF
STATISTICS?
Quantification, Variables
And Levels Of Measurement
 Rowntree (2000) distinguishes
between category variables and
quantity variables.
 Category variables can be nominal or
ordinal.
 Quantity variables can be discrete or
continuous.
Examples: Nominal Data
 Type of Bicycle
 Mountain bike, road bike, chopper, folding,BMX.
 Ethnicity
 White British, Afro-Caribbean, Asian, Chinese,
other, etc. (note problems with these
categories).
 Smoking status
 smoker, non-smoker
Ordinal Data
 A type of categorical data in which order is
important.
 Class of degree-1st class, 2:1, 2:2, 3rd class,
fail
 Degree of illness- none, mild, moderate,
acute, chronic.
 Opinion of students about stats classes-
Very unhappy, unhappy, neutral, happy,
ecstatic!
Discrete Data
Only certain values are possible (there are
gaps between the possible values). Implies
counting.

Continuous Data
Theoretically, with a fine enough
measuring device. Implies counting.
Relationships between Variables.
(Source. Rowntree 2000: 33)

Variables

Category Quantity

Continuous
Ordinal Discrete
Nominal (counting) (measuring)

Ordered
categories Ranks.
Quantification, Variables
And Levels Of Measurement
 Fielding and Gilbert (2000: 15)
distinguish between four levels of
measurement.
1. Nominal
2. Ordinal.
3. Interval
4. Ratio.
Interval and ratio variables
 According to Fielding & Gilbert (2000) these are often used
interchangeably, and incorrectly by social scientists.(pg15)
 Interval, ordered categories, no inherent concept of zero
(Clark 2004), we can calculate meaningful distance between
categories, few real examples of interval variables in social
sciences. (Fielding & Gilbert 2000:15)
 Ratio. A meaningful zero amount (eg income), possible to
calculate ratios so also has the interval property (eg someone
earning £20,000 earns twice as much as someone who earns
£10,000).(ibid)
 Difference between interval and ratio usually not important for
statistical analysis (ibid).
Interval variables- Examples
 Fahrenheit temperature scale- Zero is
arbitrary- 40 Degrees is not twice as hot as
20 degrees.
 IQ tests. No such thing as Zero IQ. 120 IQ
not twice as intelligent as 60.
 Question- Can we assume that attitudinal
data represents real, quantifiable measured
categories? (ie. That ‘very happy’ is twice as
happy as plain ‘happy’ or that ‘Very
unhappy’ means no happiness at all).
Statisticians not in agreement on this.
Ratio variables-Examples
 Can be discrete or continuous data.
 The distance between any two adjacent units of
measurement (intervals) is the same and there is
a meaningful zero point (Papadopoulos 2001)
 Income- someone earning £20,000 earns twice
as much as someone who earns £10,000.
 Height
 Unemployment rate- measured as the number of
jobseekers as a percentage of the labour force
(ibid).
IMPORTANT! SEE
‘TYPES OF DATA
REVISION SLIDES’ ON
MY WEBSITE FOR
EXTRA INFORMATION
ON TYPES OF DATA
Frequencies and Distributions
 Frequency-A frequency is the number of
times a value is observed in a distribution or
the number of times a particular event
occurs.
 Distribution-When the observed values are
arranged in order they are called a rank
order distribution or an array. Distributions
demonstrate how the frequencies of
observations are distributed across a range
of values.
 Example
Look at the distribution below:
This distribution shows the recorded ages of patients receiving
treatment for heart disease in the Stroud district. There are
50 observed values. We can easily see how often each
value occurs. What is the frequency of the following values,
79; 81; 94? What is the range of this distribution?(r = h – l ).
What is the mode? What is the median? From this
distribution we can also tell that most of the values tend to
cluster around the middle of the range.
62 64 65 66 68 70 71 71 72 72
73 74 74 74 75 75 76 77 77 78
78 78 79 79 79 80 80 80 81 81
81 81 81 82 82 82 83 83 85 85
86 87 87 88 89 90 90 92 94 96
Two elements to a distribution
 Scale with a number of values -
(Usually arrange the scores from the
highest to lowest).
 Corresponding observations- Tally up
the scores, convert them into
frequencies.
Types of Distribution
 Frequency distribution
 Class Intervals
 Relative (Proportional or percentage
distributions)
 Cumulative distributions.
Frequency Distributions
 Shows number of cases having each of the
attributes of a particular variable. Divided
into two types
1. Ungrouped distribution-scores not
collapsed into categories, each score
represented as a separate values
2. Grouped distribution. Scores collapsed into
categories so that several scores are
presented together as a group. Groups
usually referred to as a class interval.
Relative (proportional or
percentage) distributions
 The proportion of cases in the whole
distribution observed at each score or
value.
Cumulative distribution.
 The number of cases up to and
including the scale value. Can appear
in grouped or ungrouped format.
 Cumulative relative distribution for any
particular value is the the total up to,
and including, that value