Quantitative Methods

Les Oakshott

• Publisher: Bloomsbury Academic
• Year: 2014
• Print ISBN: 9781137340856
• eBook ISBN: 9781350305441
• Edition: 1
• Topic: Betriebswirtschaft
• Subtopic: Wirtschaftsmathematik

Offering a student-friendly introduction to quantitative methods, this text covers all of the need to know basics in a clear and engaging manner. Quantitative Methods is an ideal text for students of all levels coming to the subject for first time.

1

COLLECTING DATA:
SURVEYS AND SAMPLES

OBJECTIVES

• To understand the difference between a sampling frame, a sample and a population

• To understand the difference between probabilistic and non-probabilistic sampling

• To know how to select the correct sampling method in different situations

• To understand how to design a questionnaire

INTRODUCTION

Many decisions made by business and by the government are the result of information obtained from sample data, as it is often too costly or impractical to collect data for the whole population.

Data may already exist or it may need to be collected. When we have to collect our own data we call it primary data. When it already exists, as in government statistics, we call it secondary data. The collection of data can take many forms, but in this chapter we will concentrate on data that is collected by carrying out surveys. During the UK General Election in 2010 many surveys were undertaken by market research companies. The purpose of these surveys was not just to try and understand how people might vote but also to look at attitudes to issues and leaders, for example. Most of these surveys would have been conducted by telephone or face-to-face, although in some other less urgent surveys people are sent questionnaires by post or email or asked to complete a questionnaire online.

In the surveys undertaken during the UK General Election a lot of care would have gone into selecting a representative sample of the target population. The samples had to be representative of the gender, age profile and ethnicity of the population being surveyed as well as other attributes that might be considered to affect people’s voting intentions or views. The number of people being surveyed would also be calculated to ensure that a specified level of accuracy was obtained. A fuller discussion of accuracy issues when taking samples of data is given in Chapter 7.

As well as discussing ways of collecting samples of data this chapter also discusses questionnaire design because it is very important to ensure that the type and wording of questions is given proper attention if the analysis is to be of use.

CASE STUDY

British Social Attitudes survey

Every year NatCen Social Research (NatCen) undertakes a survey where it asks over 3000 people in Britain what it is like to live and work in the UK. It has been conducted annually since 1983 which makes it NatCen’s longest running survey. Respondents are chosen at random using the Post Office Postcode address file. The survey is conducted face to face and is partly paper based and partly by computer. The 2011 survey covered such areas as health, immigration, welfare, Scottish Independence and transport. The interviewer had to complete around an hour’s worth of questions and a small sample is given in Table 1.1 below. As well as the questions there are instructions for the interviewer. The sample size for each question is given on the far right of each question (all 3311), and the results of three questions are given as percentages at the end of this table.

Table 1.1 Sample questions from the 2011 British Social Attitudes survey

Q264 [SocSpnd1] * N=3311

CARD B2

Some people think that there should be more government spending on social security, while other people disagree. For each of the groups I read out please say whether you would like to see more or less government spending on them than now. Bear in mind that if you want more spending, this would probably mean that you would have to pay more taxes. If you want less spending, this would probably mean paying less taxes.

Firstly, ... READ OUT ...

benefits for unemployed people: would you like to see more or less government spending than now?

Q265 [SocSpnd2] *

N=3311

CARD B2 AGAIN

(Would you like to see more or less government spending than now on ...)

... benefits for disabled people who cannot work?

Q266 [SocSpnd3] *

N=3311

CARD B2 AGAIN

(Would you like to see more or less government spending than now on ...)

... benefits for parents who work on very low incomes?

Q267 [SocSpnd4] *

N=3311

CARD B2 AGAIN

(Would you like to see more or less government spending than now on ...)

... benefits for single parents?

Q268 [SocSpnd5] *

N=3311

CARD B2 AGAIN

(Would you like to see more or less government spending than now on ...)

.. benefits for retired people?

Q269 [SocSpnd6] *

N=3311

CARD B2 AGAIN

(Would you like to see more or less government spending than now on ...)

... benefits for people who care for those who are sick or disabled?

Source: NatCen Social Research’s British Social Attitudes survey

THE BASICS OF SAMPLING

A survey only collects information about a small subset of the population. The word ‘population’ can and often does refer to all the people in Britain or a town, but for statisticians it is also a general term used to refer to all groups or items being surveyed. For instance, it could refer to the viewing habits of all children in a town or, as you will see in a later chapter, it could refer to the weights of jars of coffee produced by a company during a week. The alternative to a survey is to question every member of the population, and when this is done it is called a census. Unfortunately it is expensive and very difficult to carry out a census, and also unnecessary. A survey of a small subset of the population, called a sample, can give surprisingly accurate results if carried out properly. This and other chapters will show you what factors must be taken into account to give accurate results.

It is crucial to be clear about the purpose of the survey as this will define your target population. Once you have selected your target population, you need to determine whether there is any list that would allow you to identify every me...

Table of contents

• Cover
• Title Page
• Copyright
• Contents
• List of Figures
• List of Tables
• Preface to Quantitative Methods
• Acknowledgements
• 1. Collecting Data: Surveys and Samples
• 2. Describing Data Using Tables and Charts
• 3. Describing Data Using Simple Statistics
• 4. Index Numbers
• 5. Applications of Probability
• 6. Probability Distributions
• 7. Analysis of Sample Data
• 8. Hypothesis Testing
• 9. Correlation and Regression
• 10. Investment Appraisal
• 11. Time Series Analysis
• 12. Linear Programming
• 13. Critical Path Analysis
• Appendix 1 Statistical Tables
• Bibliography
• Index