rm unit2_lesson3 (1)

7/30/2019 RM Unit2_Lesson3 (1)

1/14

Module: Research methods

Unit: Data gathering techniques

Lesson: Quantitative techniques

2012 Resource Development International Ltd. All rights reserved.

Resource Development International Limited reserves all rights of copyright and all other intellectual property rights in these

learning materials. No part of any learning materials may be reproduced, stored in a retrieval system or transmitted in any form

or by any means, including without limitation electronic, mechanical, photocopying, recording or otherwise, without the prior

written consent of Resource Development International Limited.


2/14

Q u a n t i t a t i v e t e c h n i q u e s

Questionnaires

Questionnaires are much the most commonly used type of quantitative investigation in business and

management. Another source is from experiments, for example measuring output levels under

different working conditions. Other quantitative data may be derived from, for example, Governmentrecords, sales records, production records and so on.

Questionnaires form part of a survey and are used extensively to collect qualitative and/or

quantitative information relating to the following types of information.

Behaviour.

Facts.

Figures.

Knowledge.

Opinions.

Attitudes.

Motivations.

Lifestyle.

The steps in the survey research process are summarised in the figure.



3/14

Figure 2.1.3 - Figure 2.1.3

The format of the questionwill influence the response so you need to be able to distinguish between

multiple choice (which seek a single answer) and open-ended questions (which encourage the

person to express feelings or expand ideas).

A typical open-ended question in a job interviewmight be 'What aspects do you like about your job?

"

List what you think the drawbacks of 'open-ended' questions are when collecting quantitative

data?



4/14

Some of the points you could have considered are:

Most respondents do not like open-ended questions because they are time consuming toanswer.

It is very difficult to control the response, as people often tend to ramble and stray from thequestion.

They are difficult to synthesise and to group into categories.

Ordering the questions

Once the questions have been drafted and the format of response selected, attention then needs to

be given to the order so that it will clearly establish the nature of the questionnaire whilst conveying

that the filling out of the questionnaire will be straight forward and will not be time consuming.

Do not begin with questions which require considerable thought or time to gather information. Make

difficult questions as easy as possible to answer. For example, where you need a specific value

such as income, it is often easier to provide a range of answers, such as, 10,000 to 15,000,

15,001 to 20,000, 20,001 to 25,000 and so on.

Progression from one question to the next should follow a logical sequence that confirms theobjectives of the questionnaire.

Other considerations

As well as the points made already, the questionnaire should also be reviewed against the following

questions. Does it:

maintain the involvement of the respondent?

develop respondent confidence?

minimise the difficulty of data collection and processing?

If the answers to these questions are satisfied, the questionnaire is ready for piloting. Trying thequestionnaire on a few respondents is likely to highlight any problems, such as questions that can

be misinterpreted or cannot be understood easily.



5/14

Administering a questionnaire

The most common ways are:

Posting - If there is an appropriate mailing list available, posting questionnaires is a usefulmeans of collecting information. However, in the absence of incentives to complete, response

rates may be poor.

Self-completion - For internal purposes, questionnaires can be delivered by hand. This puts

pressure on employees to complete, and response rates should be good. In addition,

employees are more likely to fill out the questionnaire as they see themselves as playing animportant role in influencing the outcomes of any planned changes.

Telephone questionnaires - This is a relatively low-cost means of achieving high response

rates, particularly in business-to-business research. However, to be effective, these must be

brief and utilise questions which give factual answers, so closed questions are the most

suitable.

Collect a range of questionnaires, including formal, informal and junkmail surveys.

Assess the effectiveness of each, explaining your reasoning.

The key here is to consider them in relation to their purpose.

If you have any doubts about their effectiveness, try discussing themwith colleagues or other

students.

Computer-aided interviewing and questionnaires

Computer systems make the task of gathering data much easier for the researcher. Interviews and

questionnaires can bemademuch simpler to work with if appropriate software is utilised.

In both cases, it is possible to gather data directly on to the computer by encouraging members of a

sample to respond to questions posed, not by researchers, but by the computer itself. Responses

can be set as multiple-response or bymeans of a points scoring system, i.e. scoring 0 for completedissatisfaction through to scoring 10 for total satisfaction. The computer can thus gather a series of

quantitative responses that can be analysed easily.

The major advantages of using the computer for interview and questionnaire situations are that:

the additional pressure caused by the presence of the interviewer is removed

questions can be answered at times that suit the interviewee as opposed to a set

appointment, and

some individuals feel that they can bemore honest with their answers when they are



6/14

responding to an inanimate object.

If any or all of these factors lead to more accurate information then the use of computer-aided data

gathering should be seriously considered.

Internet surveys

The availability of the Internet to a wide audience is the key factor in its importance in the data

gathering process. Data can be collected from a muchwider geographic area than throughtraditional methods and, just as importantly, with much faster response times.

An organisation seeking to carry out a survey amongst its entire global sales force could, at least in

theory, achieve this within the sameday It could also be able to gather and present the information

ready for presentation within the same period of time. Inevitably, this would require a great deal of

coordination, but it should be possible to achieve such rapid response. If the information can then

be disseminated to managers, the organisation will be in a better position to respond to marketchanges than its immediate rivals. It would be seen to have gained a potential competitive

advantage.

Look at a range of websites. How is information about potential customers gathered?

Describe four different examples.

Find examples of surveys carried out through websites.

Pilot testing

Having reviewed the structure and content of a questionnaire or survey, it is still not ready for market

release. It is necessary to carry out a pilot test or study.

This involves carrying out the questionnaire with a small selected sample to identify any problems

that could arise. For example, it is necessary to find any possible ambiguities or mistakes in the

format of the questions before it is circulated to its full sample. Checks should also be made to

replace any leading questions as these could influence individuals' responses.

The use of a pilot study is absolutely crucial when using questionnaires and ideally needs to beundertaken in two phases:

Asking colleagues their opinion.1.

Using a sample population who are part of the intended population to receive the

questionnaire.

2.

The purpose of using a sample is to examine all responses to see if the responses are clear and/or

appropriate. Adjustments would then bemade accordingly to the questions. In addition, a trial run of

the analysis of the results should be undertaken to see if the results are meaningful and useful.

Designing data collection methods for computational analysis

To ensure that all data is available in a suitable form for computational analysis it is usual to design 2012 Resource Development International Ltd. All rights reserved.


7/14

the responses so that they can be easily coded. By converting each response into a numeric, or

alpha-numeric, code it is possible to enter the data into a computer with relative ease and facilitate

the use of a suitable data analysis software package.

The simplest form of coding consists of binary values for yes and no responses. Yes is normally

allocated a value of 1, whilst a no response is allocated a 0. A series of eight responses, for example, might generate a coded value of 11010001, indicating four positive and four negative

answers.

Design a suitable simple coding system for a short series of questions about gender, age, and

nationality.

Your questions might be:

1. Select gender male/female

2. Select age group under 25, 26 to 35, 36 to 45, over 46

3. Select nationality British, French, German, other

Your responses could then be coded as follows:

Gender male = 1, female = 2

Age group under 25 = 1, 26 to 35 = 2, 36 to 45 = 3, over 45 = 4

Nationality British = 1, French = 2, German = 3, Other = 4

A response of 223 would thus represent a German female aged between 26 and 35.

In almost all cases, more detailed coding systems will be needed. These will probably includecoding of analytical categories, using appropriate abbreviations or pseudonyms. This is an area

which is outside the scope of this module and the student is advised to investigate through suitable

texts or via the Internet in order to gain a more complete understanding. One useful starting place is:

http://www.socialresearchmetho s.net/kb/statprep.php

There are a variety of software packages available which will greatly minimise the time that manualanalysis would consume. Examples include:

Microsoft Office - storing notes and other forms of data in spreadsheets, word processing anddatabases.

SPSS (Statistical Package for Social Sciences) - general statistical analysis using a range of

techniques. Equally applicable to business andmanagement research.



8/14

Ethnography - filing and indexing text into different categories. Data is analysed using an

anthropological framework.

NUD-IST (Non-numerical, Unstructured Data Indexing - Searching and Theorising) and Atlas/it

- both tools can generate new categories and relationships. They is a powerful analytical tool

allowing for analysis of key words and phrases in written data and analysis of audio recordings

and documents.

Experiments

Among the best known experiments relating to management are a series carried out in the 1920s.

The methods used during the Hawthorne experiments are just as valid today, and their conclusionshave formed the basis of many theories or management and organisational behaviour.

For an outline of the key procedures, see:

http://www.experiment-resource .com/hawthorne-effect.html

There is more extended discussion at:

http://en.wikipedia.org/wiki/H wthorne_effect

The experiments are also discussed from anHRperspective at the following website. The author

considers that the improvements in output were influenced by the fact that the researchers were

carrying out experiments, and also that workers were grouped into teams for the experiment (a

factor which itself created improvements). In both cases note that the key point is that the

experiment itself affected what it set out to measure. See:

http://www.personneltoday.com/ rticles/2004/10/19/26126/staff-under-the-microscope.html

To what extent could experiments be used in your possible research projects?

What would be the main benefits and difficulties?

Before your group meets, work individually to identify:

which forms of data collection you are likely to use in your research project

what problems and issues you expect to encounter in the data collection stage.

As a group discuss the various techniques and offer suggestions about how each student'sproblems and issues could best be overcome or dealt with. Note that this approach, in addition to

helping one student with a particular issue, is also likely to bring up new ideas that may apply to

your own research.



9/14

Knowledge Checks



12/14

Knowledge Checks - Solutions



14/14

rm unit2_lesson3 (1)

Documents