Construction of theory

Collection of data

Design of methods for gathering the data

Social Research

Empirical reality is treated as the privileged source of our theoretical

understanding of the social world –Empiricism

Vs.

Theories are treated as the privileged source of our understanding of

empirical reality - idealism

denies that one can read off theories from observations of the social world

Gilbert (2001) Researching Social Life

3

Bahamas 2.2

Greece 3.1

Italy 5.5

Spain 5.7

UK 6.3

United States 10.2

Sweden 14.4

France 15.9

USSR 17.4

Austria 18.9

Denmark 20.1

Finland 23.7

Hungary 32.0

Sri Lanka 35.8

Suicide

Rates in

Selected

Countries

4

One of the most important words in a scientist’s vocabulary

Science – from French and Latin first appeared in English in 14th

century – meaning knowledge

Now associated with specific body of knowledge or skill or a

branch of learning and in particular with theoretical and methodical

study of nature

Science is typically conceptualized as ‘an empirical enterprise in

which data are obtained and analysed in order to establish facts

about the world and to test explanations’ (Thomas, 2006)

From a positivist perspective reference to data is the “ultimate

court in which factual disagreements and theoretical disputes can

be resolved.”

An understanding of data is crucial for the researcher because of

its central role in the research process

5

Data are assumed to be given

› Researcher’s task is to collect these data and then analyse

them in various ways

› Language suggests data are simply out there waiting to be

picked up

› Researcher merely gathers them, taking care not to damage

them in the process

› E.g. survey interviewers must not ask leading questions &

observers are careful not to influence the behaviour and events

they are observing

› Idea of a theory neutral language and the separation of the

knower and the known

The Cartesian Dualism: Separation of the Subject and Object

(Thomas, A.B. (2006) Research Concepts for Management Studies. Routledge

6

Language of data construction has different connotations:

› Data are never simply there of given but are always actively

constructed as data (Thomas, 2006)

› E.g. Organizational Structures do not exist other than through

their ongoing enactment by members of the organization –

perhaps following a design created by an organizational

theorist.

› E.g. accounting numbers and other official statistics do not

occur naturally but are created by agents in particular

circumstances for particular purposes.

› Researchers do not simply report what is there but play an

active role in generating descriptions of the world and plausible

research according to conventions of scientific discourse.

7

Word data is derived from the Latin verb dare – to give and its past participle datum. A ‘piece of information’, literally ‘something given’.

Origin: Mid 17th century: Originally recorded as a term in philosophy referring to ‘things assumed to be facts or givens.

This identity with facts is retained in the modern core meaning. In the ‘standard model’ of science data are equated with observations.

Through observations there is also a link with experiment in its original meaning of experience.

Empirical data = experience data

“By implication the objects of our knowledge the things we seek to acquire knowledge of, are independent of ourselves (not imagined or supposed) and posses qualities that we may or may not be able to discover depending on the methods we use to seek knowledge of them” (Thomas, 2006:88)

Things and their properties are real and have an objective existence.

8

Fact commonly seen to refer to that which is true or real

Modern usage of the term fact or data often takes on a

strong quantitative connotation

› Data analysis until relatively recently implied statistical

manipulation of numeric information – from experiments,

surveys and observations.

› Tendency to associate data with numbers of statistics, of

measurements

Now it is common to discuss both quantitative and qualitative

data and their different forms of analysis.

9

Numerical data

advantages: it can be easy and relatively cheap to collect);

disadvantages: may not tell the whole story; may be misleading;

‘Quantitative Research’

data are usual distinguished in terms of their level of

measurement as nominal, ordinal, interval and ratio data.

References usually made to processes involving data

› E.g. data analysis, data reduction, data mining, data set, etc.

Primary distinction for data itself is that between quantitative

and qualitative data.

This distinction is problematic – notion of qualitative data is

ambiguous.

10

What do the following have in common?

Applying for a passport or applying for a place on a SHU

doctoral programme

Being stopped in the street to answer a 'few questions' about

your shopping habits, product preferences or voting

intentions

A telephone caller 'wondering' whether you would participate

in a survey regarding double glazing products for your shed.

11

The approach to collect data varies

the objective is always to secure responses from an

individual with regard to certain characteristics of interest

individuals approached are

› Units of Analysis

› Observations

› Cases or Subjects

Characteristics studied are termed Variables

Responses linking individuals to the characteristics are

known as Values

Units of analysis, variables and values = 'Data'

12

Addresses three distinct issues

› Topic of Interest - as described by the variables

› Respondents - indicated by the units of analysis

› Responses - reflect the values of the variables

Variables can assume different values for different units of

analysis

› if all units of analysis have the same value it is a Constant

Diamantopoulos, A. & Schlegelmilch, B.B. (1997) Taking the Fear Out

of Data Analysis. Dryden Press

13

According to Meaning

› Facts - characteristics or situations that exist or have existed;

descriptions of individuals' behaviour

› Awareness or knowledge of some object or phenomenon

› Intentions - acts that people have in mind to do

› Attitudes and Opinions - views, preferences, inclinations or

feelings of people towards some object or phenomenon

› Motives of individuals - internal forces that channel behaviour in

a particular way

Diamantopoulos & Schlegelmilch (1997)

14 14

Types of Secondary Data

Source: © Mark Saunders, Philip Lewis and Adrian Thornhill, 2006

15

According to Source

Primary data - collected with specific purpose in mind

› usually gathered by researcher via

surveys, experiments, observation methods

Secondary data –data which have not been expressly

gathered for the immediate study but for some other purpose

› Published statistics, annual report, database services, etc.

According to Time Dimension

Cross-sectional Data – single point in time

Longitudinal data –relating to a number of time periods

Trend data Diamantopoulos & Schlegelmilch (1997)

16

Which of these films have you seen?

Gone with the wind Yes No

Casablanca Yes No

Some like it hot Yes No

Dumbo Yes No

What is your highest level of

educational attainment?

GCSE

A Level

HNC/HND

Degree

PG diploma

Higher degree

NOMINAL

QUESTIONS

How often during this session

have you felt stressed?

Always

Very often

Fairly often

Sometimes

Almost never

ORDINAL QUESTIONS

17

Dumbo was an excellent film

Strongly Neither agree Strongly

agree Agree nor disagree Disagree disagree

5 4 3 2 1

David Cameron’s period as Prime Minister has been;

Excellent 1 2 3 4 5 6 7 Appalling

Data

Categorical Quantifiable

Descriptive Discrete Continuous Ranked

Increasing precision

Saunders et al, 2006

Interval Data

18

Less likely to be conceptualized as variables– aim is to

understand the meanings of human actions

Typically data takes the form of texts or meanings derived

from textual sources

› Field notes or Interview transcripts

› Research diaries, visual materials, etc.

Qualitative data are not represented numerically and are not

analysed using statistical analytical processes

Data are categorized and examined for patterns and themes

that can be integrated into narrative accounts

19

The action of

variable A Causes / explains Change in

variable B

Positivist theorists

Interpretivism

Experience &

interpretation Stimulus Response

OR Interpretation & meaning Action

Gill & Johnson, 2010

Methodological Monism

20

stimuli external

to, but

impacting upon

the actor

Inter-subjective

interpretation of

external stimuli by

the actor

Actor's rational

selection &

construction of

particular

courses of action

Norms, beliefs, values deriving

from the various cultural

contexts which the actor has

been exposed to

VERSTEHEN

Mode 1: Mode of interpretive, or emic, analysis using verstehen

Gill & Johnson, 2010

21

Epistemology Theoretical Perspectives Methodology Methods

Objectivism

Constructivism

Subjectivism

Positivism ( &post positivism)

Interpretivism

symbolic interactionism

phenomenology

hermeneutics

Critical inquiry

Feminism

Postmodernism

etc.

Experimental research

Survey Research

Ethnography

Phenomenological research

Grounded theory

Heuristic inquiry

Action research

Discourse analysis

etc.

Sampling

Measurement & scaling

Questionnaire

Observation participant, non-participant

Interview

Focus group

Case study

Document analysis

Interpretative methods

Content analysis

etc.

source: Crotty, M. (1998) The Foundations of Social Research. Sage

22 22

Traditional role of scientific method (Bacon)

1. Observation & experiment; 2. Inductive

generalizations; 3. Hypotheses; 4. Attempted

verification of hypotheses; 5. Proof or disproof; 6.

Knowledge;

Popper’s view

1. Problem (rebuff to existing theory or expectation);

2. Proposed solution (new theory); 3. Deduction of

testable propositions from new theory;

4. Tests (attempted refutations through e.g. experiments

and observations)

5. Preference established between competing theories

23

• For successful communication need agreement on meaning of words

Words fixed no fixed meanings determinism pragmatism postmodernism

• Determinism & Language

• Meaning is fixed absolutely – meaning of word determined by the object it represents: there is an intimate connection between words and things

• Establish what a word means by ostensive definition • Point to an instance of the object that the word stands for • Meaning of a word corresponds to whatever it is in the world

it represents

Alan Thomas (2006) Research Concepts For Management Studies. Routledge

24

• Postmodernism & Language • Meanings are fluid & unfixed with no possibility of

anchoring hem • Determinism is connected to modernism • Modernism concerned with boundedness, framing, certainty &

knowing what a term really means • Postmodernsim / poststructuralism reject quest fro absolute

certainty and ‘truth’ • Language has no fixed meaning – can be autonomous

phenomenon with no connection to externality, no representational function

• Language is not at disposition of an autonomous self that uses it to express truths

• Fixing meanings beyond question is impossible – words are like bubbles floating free of world & cannot be captured

Alan Thomas (2006) Research Concepts For Management Studies. Routledge

25

• Pragmatism & Language • Meanings can be stabilized for all practical purposes even

though there is no absolute basis for whatever meanings

are current

• Meanings cannot be absolutely fixed nor understood as completely fluid

• Pragmatists focus on way language is used for everday purposes of living

• Words have no transcendent meaning, however, meanings are fixed for the

‘time-being’

• Term’s meaning can be understood by situating it within the contexts of its use

• Words are meaningful by reference to the contexts of their use (e.g. goal)

• “ostensive definition does not connect a word to a unique referent” – cat has

many meanings” can’t expect terms to have absolute meaning. But words do

convey meaning & in order to understand them must locate tehm in their

contexts (p.17)

Alan Thomas (2006) Research Concepts For Management Studies. Routledge

26

The concept of data - ambiguous and elusive

uncertainty (especially in social sciences) as to where data (is) are located 'ontologically'

Difficult philosophical questions- considerations of the 'real' and our relation to it

So, 'What are data?'

Conventional usage - distinguish between the world (the source of data) and textual records of features of the world (the data)

Much qualitative 'research' involves 'ambiguity' when text (an aspect of that world) is used itself as a source of data.

Texts are represented and described by other texts - relativism?

Positivism world is assumed to be directly accessible to the senses - empiricism -facts speak for themselves

problem arises when we encounter non-observables

Social constructionism -data are never 'innocent' of presuppositions - they are constructed both by the researcher and those whom the researcher studies.