Theory of Planned Behavaior Manual-Francis[2]

Theory of Planned Behavaior Manual-Francis[2].pdf

Conduct the Point-of-Care Research Questionnaire

Theory of Planned Behavaior Manual-Francis[2]

OMB: 2900-0796

Document [pdf]

Download: pdf | pdf

CONSTRUCTING QUESTIONNAIRES BASED ON
THE THEORY OF PLANNED BEHAVIOUR

A MANUAL for HEALTH SERVICES RESEARCHERS

Authors: Jillian J Francis, Martin P Eccles, Marie Johnston, Anne Walker,
Jeremy Grimshaw, Robbie Foy, Eileen F S Kaner, Liz Smith, Debbie Bonetti

Centre for Health Services Research
University of Newcastle
21 Claremont Place
Newcastle upon Tyne
NE2 4AA
United Kingdom
May 2004
ISBN: 0-9540161-5-7
Funded by the European Union:
Contract number QLG4-CT-2002-00657

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

FOREWORD

This manual is a response to a request from health services researchers wishing to predict and
understand behaviour, in particular, researchers throughout the European Union involved in the
ReBEQI project (Research-Based Education and Quality Improvement). It is based on a psychological
model of behaviour change, the Theory of Planned Behaviour (TPB; Ajzen, 1988), which evolved from
the Theory of Reasoned Action (Fishbein, 1967) and is designed to assist psychologists and nonpsychologists involved in health services research to produce an effective questionnaire to measure the
TPB constructs. Advice from the TPB literature (e.g. Ajzen, 1988; Conner & Sparks, 1995; Godin &
Kok, 1996) has been integrated, resulting in a guide to writing questionnaires that is based on current
practice among TPB researchers. Questionnaires based on the TPB can be used to investigate the
attitudes and beliefs underlying health-related behaviour. In implementation (or knowledge transfer)
research with health care professionals, these questionnaires have been used to investigate the uptake of
evidence-based practice.
The Theory of Planned Behaviour is the explicit theoretical basis for 222 studies published in the
Medline database, and 610 studies published in the PsycINFO database, from 1985 to January 2004.
Apparently, hundreds of researchers have used this theory and have gone about the fascinating but time
consuming task of reading the source materials and considering current debates about measurement
strategies, before constructing a questionnaire to investigate their topic of interest. This manual is a tool
that may help researchers to fast-track through this process. For those who have the time to investigate
the background issues, we have included a full discussion of our operational decisions in Appendix C.
This manual has been subjected to a wide range of reviewing and trialling procedures. We are grateful
to many colleagues who reviewed earlier drafts, to workshop participants who used the manual as an
‘instant’ guide to writing questions and to our bilingual colleagues who provided translations, backtranslations and comments on the Dutch and Italian versions of the brief questionnaire.

Jill Francis, Martin Eccles, Marie Johnston, Anne Walker, Jeremy Grimshaw, Robbie Foy,
Eileen Kaner, Liz Smith, Debbie Bonetti
May, 2004

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

CONTENTS

FOREWORD ............................................................................................................................ 2
CONTENTS .............................................................................................................................. 3
1

BRIEF SUMMARY OF THE THEORY OF PLANNED BEHAVIOUR .............................. 7

HOW THE THEORY WORKS .......................................................................................... 8
2.1

Behaviour .............................................................................................................................................................. 8

2.2

Intention................................................................................................................................................................ 8

2.3

Attitudes (towards the behaviour)........................................................................................................................ 9

2.4

Subjective norms (about the behaviour)............................................................................................................. 9

2.5

Perceived behavioural control (of the behaviour) ............................................................................................. 9

2.6

Direct measures and indirect (belief-based) measures...................................................................................... 9

2.7

Reliability .............................................................................................................................................................. 9

STEPS IN THE CONSTRUCTION OF A TPB QUESTIONNAIRE ................................. 10

MEASURING BEHAVIOURAL INTENTIONS................................................................ 11

4.1

Method 1: Intention performance..................................................................................................................... 11
4.1.1 Procedure....................................................................................................................................... 11
4.1.2 Scoring .......................................................................................................................................... 11

4.2

Method 2: Generalised intention....................................................................................................................... 11
4.2.1 Procedure....................................................................................................................................... 11
4.2.2 Scoring .......................................................................................................................................... 12

4.3

Method 3: Intention simulation......................................................................................................................... 12
4.3.1 Stages of Development.................................................................................................................. 12
4.3.2 Procedure....................................................................................................................................... 12
4.3.3 Scoring .......................................................................................................................................... 12

MEASURING ATTITUDES ............................................................................................ 13
5.1

Direct measurement of attitude......................................................................................................................... 13
5.1.1 Procedure....................................................................................................................................... 13
5.1.2 Scoring .......................................................................................................................................... 13

5.2

Indirect measurement of attitude: measuring behavioural beliefs and outcome evaluations ..................... 14
5.2.1 Stages of Development.................................................................................................................. 14
5.2.2 Procedure....................................................................................................................................... 14
5.2.3 Scoring .......................................................................................................................................... 15

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

MEASURING SUBJECTIVE NORMS ............................................................................ 17
6.1

Direct measurement of subjective norm........................................................................................................... 17
6.1.1 Procedure....................................................................................................................................... 17
6.1.2 Scoring .......................................................................................................................................... 17

6.2

Indirect measurement of subjective norm: measuring normative beliefs and motivation to comply......... 18
6.2.1 Stages of Development.................................................................................................................. 18
6.2.2 Procedure....................................................................................................................................... 18
• Pilot test these items by asking about five respondents to answer the questions and tell you whether they
have any difficulty answering them. If necessary, modify the wording of the questions........................... 19
6.2.3 Scoring .......................................................................................................................................... 20

MEASURING PERCEIVED BEHAVIOURAL CONTROL .............................................. 21
7.1

Direct measurement of perceived behavioural control (PBC)........................................................................ 21
7.1.1 Procedure....................................................................................................................................... 21
7.1.2 Scoring .......................................................................................................................................... 22

7.2

Indirect measures of PBC: Measuring control beliefs and their perceived power to influence behaviour 22
7.2.1 Stages of Development.................................................................................................................. 22
7.2.2 Procedure....................................................................................................................................... 22
7.2.3 Scoring .......................................................................................................................................... 23

STEPS IN MANAGING A TPB SURVEY ....................................................................... 25
8.1

Defining the population, clinical condition and behaviour of interest........................................................... 25

8.2

Elicitation Study ................................................................................................................................................. 25
8.2.1 Collecting the data......................................................................................................................... 25
8.2.2 Analysing the data ......................................................................................................................... 25
8.2.3 Wording the questions................................................................................................................... 25

8.3

Formatting and piloting the questionnaire ...................................................................................................... 26
8.3.1 Number and content of questions .................................................................................................. 26
8.3.2 Ordering of questions in the document.......................................................................................... 26
8.3.3 Piloting the questionnaire and creating the second draft ............................................................... 27

8.4

Brief forms of the questionnaire ....................................................................................................................... 27

8.5

Translating questionnaires into other languages............................................................................................. 28

8.6

Main study: Practical issues .............................................................................................................................. 29
8.6.1 What sample size is needed? ......................................................................................................... 29
8.6.2 Other aspects of survey methodology that need to be kept in mind .............................................. 29

DATA ENTRY AND ANALYSIS .................................................................................... 30
9.1

Setting up the data file ....................................................................................................................................... 30

9.2

Data screening .................................................................................................................................................... 30

9.3

Main Analysis ..................................................................................................................................................... 30

REFERENCES AND SOURCES ................................................................................... 31

GLOSSARY ................................................................................................................... 32
4

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

APPENDIX A: QUESTIONNAIRE DEVELOPMENT MATERIALS ............................... 35
12.1 Example, Elicitation Study Materials............................................................................................................... 35
12.2 Example, Questionnaire..................................................................................................................................... 36
12.3 Example, scoring key for questionnaire ........................................................................................................... 39

13 APPENDIX B: EXAMPLES OF BRIEF QUESTIONNAIRES IN LANGUAGES OTHER
THAN ENGLISH ..................................................................................................................... 41
13.1 Dutch language ................................................................................................................................................... 41
13.2 Italian language .................................................................................................................................................. 42

APPENDIX C: DISCUSSION PAPER ...................ERROR! BOOKMARK NOT DEFINED.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

BRIEF SUMMARY OF THE THEORY OF PLANNED BEHAVIOUR

To predict whether a person intends to do something, we need to know:
•
•
•

Whether the person is in favour of doing it (‘attitude’)
How much the person feels social pressure to do it (‘subjective norm’)
Whether the person feels in control of the action in question (‘perceived behavioural control’)

By changing these three ‘predictors’, we can increase the chance that the person will intend to do a desired
action and thus increase the chance of the person actually doing it. In a clinical consultation, the clinician’s
treatment decisions and actions are examples of intentional behaviour.
In implementation research, clinical guidelines make evidence-based recommendations about the actions of
clinicians in order to maximise quality of care. Many interventions are likely to work through clinicians’
attitudes, subjective norms and perceived behavioural control – enhancing these is likely to increase
compliance with guidelines.
Box 1.1
A patient with atrial fibrillation presents to their GP for an annual review. The patient is not currently being
treated with warfarin. Will the GP prescribe warfarin? The answer to this depends on whether the GP intends to
do so. In other words, it is not an automatic, habitual or thoughtless action. The intention, in turn, depends on:•

Whether, overall, the GP has a positive or negative attitude to prescribing warfarin for patients with
atrial fibrillation

•

To what extent the GP perceives that they experience social pressure to prescribe or
not, including whether the GP thinks that
the patient wants warfarin
professional colleagues would approve of prescribing;
the health care system encourages prescribing
- and how important these various people’s opinions are to the GP

•

Whether the GP finds it difficult to prescribe i.e. how difficult it is to enact the behaviour in the given
context.

The Theory of Planned Behaviour can be useful in designing strategies to help people to adopt healthy
behaviours and to help clinicians increase their uptake of guidelines. This manual is a guide about how to
construct and score a questionnaire which measures the variables in this model. The relevant literature
contains some controversies and debates about how best to do this, but this document does not explore
these; we have made decisions about solutions (justified in Appendix C). The focus here is on assisting
researchers to construct a theory-based research tool in a systematic and replicable manner.
Construction of these questionnaires involves the use of both qualitative and quantitative methods. The
manual is organised in nine sections, of which this is the first. The second section presents a statement of the
theory. The third lists the steps in the construction of a questionnaire based on the TPB. The fourth concerns
the measurement of intentions. Sections five to seven are about how to measure the three predictor variables:
attitudes; subjective norms; and perceived behavioural control. The eighth section is a guide to the steps
involved in managing the project, including how to construct the questionnaire as a whole. The final section,
nine, gives details of recommended approaches to analysis. Examples of materials for the preliminary
elicitation study, and construction and scoring of the questionnaire, are included in Appendix A.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

HOW THE THEORY WORKS

The Theory of Planned Behaviour (TPB; Ajzen, 1988, 1991) proposes a model about how human action
is guided. It predicts the occurrence of a specific behaviour provided that the behaviour is intentional.
The model is depicted in Figure 1 and represents the three variables1 which the theory suggests will
predict the intention to perform a behaviour. Intentions are the precursors of behaviour.
ATTITUDES
(Behavioural beliefs x
Outcome evaluations)

SUBJECTIVE NORMS
(Normative beliefs x
Motivation to comply)

BEHAVIOURAL
INTENTIONS

BEHAVIOUR

PERCEIVED
BEHAVIOURAL
CONTROL
(Control beliefs x influence
of control beliefs)

Figure 1. The Theory of Planned Behaviour (Ajzen, 1991)

The variable names in this model reflect psychological constructs and so they have a special meaning
within the theory. Here are some brief explanations of these special meanings:
2.1 Behaviour
In implementation research, interventions are designed to change the behaviour of clinicians. The target
behaviour should be defined carefully in terms of its Target, Action, Context and Time (TACT). For
example, consider the behaviour, ‘referring patients with back pain for a lumbo-sacral spine x-ray’.
Here the target is the patient, the action is the referral, the context is the clinical condition (back pain)
and the time is (implicitly) during the consultation.
This manual is about the application of the TPB to specific behaviours. It is possible to use the model to
investigate more general behaviours (e.g. ‘x-ray referral behaviour’), but it is important to observe the
‘principle of compatibility’ (Fishbein, 1967). That is, all the measures in the questionnaire – including
the measure of intention – should refer to the same level of generality. For example, a questionnaire that
includes a measure of attitudes towards prescribing antibiotics in general and a measure of intentions to
prescribe antibiotics for uncomplicated sore throat violates the principle of compatibility. Any low
correlations could be due to this as much as any other problems with the theory or its operationalisation.
2.2 Intention
Although there is not a perfect relationship between behavioural intention and actual behaviour,
intention can be used as a proximal measure of behaviour. This observation was one of the most
important contributions of the TPB model in comparison with previous models of the attitude-behaviour
relationship. Thus, the variables in this model can be used to determine the effectiveness of
implementation interventions even if there is not a readily available measure of actual behaviour.
1

Note that the three variables also influence one another. Although Figure 1 is presented in a simplified form, a more
detailed diagram would include double-ended arrows joining these three variables.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

2.3 Attitudes (towards the behaviour)
Attitude toward the behaviour is a person’s overall evaluation of the behaviour. It is assumed to have
two components which work together: beliefs about consequences of the behaviour (behavioural
beliefs; e.g. ‘referring the patient for an x-ray will decrease future consultations’) and the
corresponding positive or negative judgements about each these features of the behaviour (outcome
evaluations; e.g. ‘decreasing future consultations is … desirable/undesirable’).
2.4 Subjective norms (about the behaviour)
Subjective norms are a person’s own estimate of the social pressure to perform or not perform the
target behaviour. Subjective norms are assumed to have two components which work in interaction:
beliefs about how other people, who may be in some way important to the person, would like them to
behave (normative beliefs), e.g. ‘I feel pressure from patients to refer them for an x-ray’) and the
positive or negative judgements about each belief (outcome evaluations), e.g. ‘in regard to my
decision to x-ray, doing what patients think I should do is important/ unimportant’).
2.5 Perceived behavioural control (of the behaviour)
Perceived behavioural control is the extent to which a person feels able to enact the behaviour. It has
two aspects: how much a person has control over the behaviour (e.g. low control over measuring
blood pressure if the BP machine often malfunctions); and how confident a person feels about being
able to perform or not perform the behaviour (e.g. not sufficiently skilled in measuring blood
pressure). It is determined by control beliefs about the power of both situational and internal factors to
inhibit or facilitate the performing of the behaviour (e.g. ‘Whether I measure a patient’s blood
pressure is entirely up to me’; ‘I could measure my patient’s blood pressure if I wanted to’).
2.6 Direct measures and indirect (belief-based) measures
With the exception of behaviour, the variables in the TPB model are psychological (internal)
constructs. Each predictor variable may be measured directly e.g. by asking respondents about their
overall attitude, or indirectly e.g. by asking respondents about specific behavioural beliefs and
outcome evaluations. Direct and indirect measurement approaches make different assumptions about
the underlying cognitive structures (see Appendix C) and neither approach is perfect. When different
methods are tapping the same construct, scores are expected to be positively correlated, so it is
recommended that both be included in TPB questionnaires (but see Section 8.4 about brief forms of
the questionnaire). This manual explains how to construct questions for both types of measure.
2.7 Reliability
It is important to establish the reliability (Everitt, 1996) of each measure. For direct measures, one
form of reliability may be established using an index of internal consistency (to determine whether the
items in the scale are measuring the same construct). However, because people can quite logically
hold both positive and negative beliefs about the same behaviour, it is not appropriate to assess the
reliability of indirect measures using an internal consistency criterion. For example, someone may
believe that referring patients with back pain will reassure the patients and also that an x-ray will
expose these patients to unnecessary radiation. Equally, a GP may be highly motivated to comply with
the expectations of professional colleagues but not at all motivated to comply with the expectations of
patients. Hence, it does not make sense to eliminate some of these beliefs from overall measures on
the grounds of low or negative correlations among them. It is necessary to use test-retest reliability (or
‘temporal stability’) for this purpose.
9

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

3 STEPS IN THE CONSTRUCTION OF A TPB QUESTIONNAIRE
The construction of a questionnaire to measure the variables in the TPB model proceeds in nine phases,
some of which involve short but important tasks, with others involving a long process of empirical
investigation. These steps are:
(a) Define the population of interest (examples: General medical practitioners in the Tayside region
of Scotland; General dental practitioners in Newcastle, Durham and Northumberland, in the
north-east of England). Decide how best to select a representative sample from this population.
(b) Carefully define the behaviour under study (explained using the TACT principle in Section 2.1
above). Use this definition to construct a general introductory statement for the start of the
questionnaire, for example, “Each question in this section refers to MEASURING THE BLOOD
PRESSURE (BP) of your patients with Type 2 diabetes”. (See Appendix A.)
(c) Decide how best to measure intentions (explained in Section 4).
(d) Determine the most frequently perceived advantages and disadvantages of performing the
behaviour (explained in Section 5).
(e) Determine the most important people or groups of people who would approve or disapprove of
the behaviour (explained in Section 6).
(f) Determine the perceived barriers or facilitating factors which could make it easier or more
difficult to adopt the behaviour (explained in Section 7).
(g) For a standard TPB-based study, include items to measure ALL of these constructs in the first
draft of the questionnaire (see example in Appendix A).
(h) Pilot test the draft and reword items if necessary.
(i) Assess the test-retest reliability of the indirect measures by administering the questionnaire
twice to the same group of people, with an interval of at least two weeks.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

4 MEASURING BEHAVIOURAL INTENTIONS
Key:

iii
ii
i

very time consuming (allow weeks)
quite time consuming (allow days)
not very time consuming (allow hours)

This section describes three methods of measuring intentions. We have called Method 1 Intention
Performance, because in some situations, it would be possible to observe actual performance using the
same measurement scale, and this direct comparability could be useful for some studies. In the TPB
literature, where most research has been about individual’s own health-related behaviour (e.g. smoking,
exercise), Generalised Intention (Method 2) is most commonly used. When investigating the behaviour
of health care professionals, Intention Simulation (Method 3) could be a more valid proxy measure for
actual behaviour, because it more closely approximates ‘real’ situations that require complex clinical
decisions. However, it is time consuming and should be prepared with great care, or it may be
misleading. A thorough conceptual analysis and review of Intention Simulation, carrying the
appropriate warnings, is presented by Jones, Gerrity and Earp (1990).
In general, the methods used to measure intentions should be guided by researchers’ judgements about
which types of questions seem to make sense for the behaviour and sample under investigation.
4.1
i

Method 1: Intention performance

4.1.1 Procedure
Use a single item in the format shown in Box 4.1
Box 4.1
Given 10 patients presenting with back pain for the first time, how many patients would you expect to
refer for an x-ray?
0 1 2 3 4 5 6 7 8 9 10

4.1.2 Scoring
For this single item measure, the number circled is the behavioural intention score.
4.2
i

Method 2: Generalised intention

4.2.1 Procedure
Adequate internal consistency can be demonstrated using three items. We suggest using the format
shown in Box 4.22.
Box 4.2
1. I expect to refer patients with lower back pain for an x-ray
Strongly disagree 1 2 3 4 5 6 7 Strongly agree
2. I want to refer patients with lower back pain for an x-ray
Strongly disagree 1 2 3 4 5 6 7 Strongly agree
3. I intend to refer patients with lower back pain for an x-ray
Strongly disagree 1 2 3 4 5 6 7 Strongly agree
2

There is a research literature that discusses the conceptual differences between these three questions, although empirically,
there is very considerable response consistency between these items. See Armitage and Conner (2001) for a discussion of
this issue.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

4.2.2 Scoring
Calculate the mean of the three intention scores.
4.3

Method 3: Intention simulation

4.3.1

Stages of Development

A. i i Write scenarios
B. i
For each scenario, ask for a treatment decision
4.3.2

Procedure

A. Write 10 scenarios, of around 80 to 100 words each, describing patients presenting with the
clinical condition, based an exploratory study and/or your own or colleagues’ clinical
experience. It is important to cover a wide range of patient types and consultation situations. If
sample size permits, you may want to systematically vary the features of the scenarios e.g. age
of patient; prior consultation history; patient’s current anxiety level; patient’s preference
regarding target behaviour.
B. For each scenario, ask respondents for a treatment decision which permits a ‘Yes/No’ answer
(see Box 4.3.) You may also want to ask how difficult the decision was.
Box 4.3
The first patient is a 29 year old woman, who comes into your surgery, limping and holding her back. She
describes her back pain as sharp and unceasing. She has had it for the last 3 weeks, and for the last week she
has been unable to go to work because she can’t drive. She is very worried that whatever is causing her
problem is getting worse. She has had no previous episodes. Her last six attendances have been for repeat
prescriptions for oral contraceptives.

Your decision: Refer for x-ray?

YES

On the scale 1 to 7, how difficult was it for you to make a decision for this scenario?
Not at all
Difficult

4.3.3

7 Extremely
Difficult

Scoring

Count the number of ‘Yes’ answers. This number is the score for behavioural simulation. The higher the
number, the stronger is the intention to perform the behaviour. For decision difficulty, you may wish to
calculate the mean of responses for each participant (which may modify the relationship between
intentions and actual behaviour) or the mean for all participants across each scenario (which may reflect
differences between scenarios). The decision whether and how to use difficulty scores depends on your
research question.3
3

Decision difficulty is not part of the TPB model. We acknowledge that there is a large literature on decision difficulty and
anyone who wants to include this as a central part of their research would be wise to consult this body of work.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

MEASURING ATTITUDES
Key:

iii
ii
i

very time consuming (allow weeks)
quite time consuming (allow days)
not very time consuming (allow hours)

5.1 Direct measurement of attitude
i
5.1.1

Procedure

•

Direct measurement involves the use of bipolar adjectives (i.e. pairs of opposites) which are
evaluative (e.g. good – bad).

•

Ideally, use about four items following a single ‘stem’ which defines the behaviour under
investigation (see Box 5.1)4.

•

Include instrumental items (whether the behaviour achieves something e.g. useful–worthless)
and experiential items (how it feels to perform the behaviour e.g. pleasant – unpleasant)5.

•

Include the good – bad scale if it is appropriate to the topic, as it captures overall evaluation.

•

Arrange the items so that the ends of the scales are a mix of positive and negative endpoints.6
(See Box 5.1)

Box 5.17

Example: A patient presents with lower back pain. The target behaviour is
referring the patient for x-ray.
negative
stem

Referring a patient with acute lower back pain for x-ray is
harmful
good
pleasant (for me)
worthless

5.1.2

1
1
1
1

2
2
2
2

3
3
3
3

4
4
4
4

5
5
5
5

6
6
6
6

endpoints

7 beneficial
7 bad
7 unpleasant (for me)
7 useful

Scoring

•

Recode the items that have negatively worded endpoints on the right, so that higher numbers
then always reflect a positive attitude to the target behaviour (e.g. for ‘pleasant – unpleasant’, an
answer of 6 becomes score of 2; a score of 4 remains a 4).

•

It is important that the attitude items have high internal consistency, i.e. that scores on these
items correlate highly with each other (see Section 9.3). You may decide to omit items from the
scale to improve internal consistency.

Box 5.1 contains sample items that may not be applicable for every study. Indeed, indiscriminate use may result in low
internal consistency of the scale and hence low correlations with intentions. (Valois & Godin, 1991). Researchers are
advised to use careful pilot testing to establish whether items are appropriate.
5
What needs to be assessed here is the perspective of the actor – not the target of the behaviour. For example, in Box 5.1,
whether the behaviour is pleasant or unpleasant for the patient is more likely to emerge as a control belief (see Section 7.2.2)
but is not an item that directly measures attitude.
6
This mixing of positive and negative endpoints is a common practice that is designed to minimise the risk of ‘response set’,
or a tendency to answer questionnaire items in the same way regardless of their content. However, some scholars (e.g.
McColl et al, 2001) argue that this could be counter productive. This is a matter of judgement, in that different types of
samples may respond differently to mixed or unmixed endpoints. It would be an important question to consider when a draft
of the questionnaire is subjected to pilot testing (see Section 8.3.3).
7
A 7-option response format most often recommended in the TPB literature, although 5-option formats are sometimes used.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

•

Calculate the mean of the item scores to give an overall attitude score.

5.2 Indirect measurement of attitude: measuring behavioural beliefs and outcome evaluations
5.2.1

Stages of Development

iii

B.
C.

ii
i

5.2.2
A.

Conduct an elicitation study to elicit commonly held beliefs: Identify the content of
behavioural beliefs that are shared by the target population.
Construct questionnaire items to assess the strength of behavioural beliefs.
Construct questionnaire items to assess outcome evaluations.

Procedure
Conduct an elicitation study
•

Take a sample (about 25 people8) from the population from which you will select
respondents for the questionnaire study.

•

Use open-ended questions. These are normally presented in one-to-one interviews, but
could also be in focus group or questionnaire form. Give participants a few minutes to
list their thoughts in response to structured questions (Box 5.2).

•

Content analyse the responses into themes (behavioural beliefs) and label the themes
extracted. To increase the validity of the analysis, at least two researchers should do this
independently. List the themes in order, from most frequently mentioned to least
frequently mentioned.

Box 5.2

Example: Your patient has Type 2 diabetes. The target behaviour is taking the
patient’s blood pressure. Please take a few minutes to list your thoughts about
the following questions:

What do you believe are the advantages of measuring the patient’s blood pressure?
What do you believe are the disadvantages of measuring the patient’s blood pressure?
Is there anything else you associate with measuring the patient’s blood pressure?

Construct questionnaire items to assess the strength of behavioural beliefs
•

Select the behavioural beliefs most often listed and convert these into a set of statements.
These statements should reflect the beliefs which might affect the behaviour of the target
population (see Box 5.3 for question and response formats). Inclusion of 75% of all
beliefs stated should give adequate coverage of the belief ‘population’.

•

Pilot test these items by asking about five people from the relevant population to answer
the questions and tell you whether they have any difficulty answering them. Check
comprehension and clarity. If necessary, modify the wording of the questions.

This is the sample size recommended by Godin and Kok (1996). The sample size could legitimately be adjusted according
to the principles of qualitative research methods, such as purposive sampling that adequately represents subsets of the
population and sampling until data saturation is achieved. For these and other issues to do with qualitative research methods,
a methods text such as Seale (1998) is recommended.

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

Box 5.3.1
Question format, behavioural beliefs

Box 5.3.2
Response format, behavioural beliefs9

If I measure blood pressure (BP), I will feel that
I am doing something positive for the patient.

Unlikely

Likely

It causes a lot of worry and concern for the
patient if they are found to have high BP.

Unlikely

Likely

If I measure BP, I will detect any problems at an
early stage.

Unlikely

Likely

If I measure BP, I’ve got to see some patients
more often.

Unlikely

Likely

c
d

Construct questionnaire items to assess outcome evaluations

Convert each of the belief statements10 into the form of an incomplete sentence. By completing the
sentence (using the set response format), the participant expresses a positive or negative evaluation
of the belief statement. (See Box 5.4 for question and response formats.)
Box 5.4.1
Question format, outcome
evaluations

Box 5.4.2
Response format, outcome evaluations11

Doing something positive for
the patient is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Causing a lot of worry and
concern for the patient is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Detecting problems for these
patients at an early stage is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Having to see some patients
more often is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

•
5.2.3

Pilot test these items by asking about five respondents to answer the questions and tell you
whether they have any difficulty answering them. If necessary, modify the wording.
Scoring

For each behavioural belief, the belief score on the unlikely-likely scale is multiplied by the relevant
evaluation score on the extremely bad/extremely good scale (see Box 5.5)12. The resulting products
across are summed all the beliefs to create an overall attitude score:
9

Some researchers prefer the endpoints ‘Strongly disagree; strongly agree’. These may make more sense for some items and
we recommend that researchers use their own judgement. We have used ‘unlikely’ and ‘likely’ in the example because an
important feature of these items is that the respondent is making a judgement about the probability that a given item is true.
10
Some outcome evaluation questions seem odd to participants. For example, asking whether ‘doing something positive for
the patient’ is ‘undesirable’ seems downright silly, and this may affect response rates or response validity. For this and other
items in which a pilot study indicates there will be zero variance, it is worth considering omitting the item from the
questionnaire and weighting the behavioural belief score by a constant selected intelligently by the researcher.
11
Some researchers prefer the endpoints ‘extremely unimportant; extremely important’. We suggest ‘extremely undesirable;
extremely desirable’ as they are directional, thereby enabling the respondent to say what are positive and negative items. For
example, the researcher may think that ‘causing worry and concern’ is a negatively worded item, but a clinician may feel
that a small amount of worry is good for the patient as it may increase the chances of compliance with treatment regimens.
12
Note that this form of scoring for attitudes and for the other predictor variables is somewhat controversial. Our reasons for
recommending this approach are given in a discussion paper (Appendix C). Briefly, we think that interpretation of scores is
easier when the midpoint of the scale is zero, so it is clear whether the final score represents an influence for or against
enacting the behaviour. Also, response scales are unipolar (1 to 7) or bipolar (-3 to +3), depending on whether the concept to
be measured is unidirectional (e.g. probability) or bidirectional (e.g. evaluation).

ReBEQI WP2 Theory of Planned Behaviour Questionnaires: Manual for Researchers

A = (a x e) + (b x f) + (c x g) + (d x h)

Formula 5.1
Where

A = total attitude score
a, b, c and d are scores for each of four behavioural beliefs
e, f, g and h are scores for outcome evaluations relating to each behavioural belief

Using this method,
•
•

a positive (+) score means that, overall, the participant is in favour of taking blood pressure
readings of patients with Type 2 diabetes.
a negative (-) score means that, overall, the participant is against taking blood pressure readings
of patients with Type 2 diabetes.

Box 5.5 Example, scoring procedure
a

C
D

If I measure blood pressure (BP), I
will feel that I am doing something
positive for the patient.

Unlikely

Likely

It causes a lot of worry and concern
for the patient if they are found to
have high BP

Unlikely

Likely

If I measure BP, I will detect any
problems at an early stage.

Unlikely

Likely

If I measure BP, I’ve got to see
some patients more often.

Unlikely

Likely

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Doing something positive for the
patient is:

Causing a lot of worry and concern
for the patient is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Detecting problems for these
patients at an early stage is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Having to see some patients more
often is:

Extremely
undesirable

-3

-2

-1

Extremely
desirable

Imagine that a participant has responded by circling the numbers indicated in bolded italics above.
The total attitude score is calculated as
A = (5 x +3) + (2 x -2) + (6 x +3) + (2 x -1)
= (+15) +(-4) + (+18) +(-2)
= +27
Because there are 4 items, the possible range of total scores is (7 x +3) x 4 = -84 to +8413
THEREFORE, THE ATTITUDE SCORE OF THE PARTICIPANT REFLECTS A WEAK TO
MODERATE POSITIVE ATTITUDE (i.e. IN FAVOUR OF MEASURING BLOOD PRESSURE)

The range will of course differ if different numbers of items are used, so interpretation of weak, moderate and strong
attitudes will be determined by the possible range. The important aspect of this measurement scheme is that zero represents a
neutral attitude, positive scores represent attitudes in favour of the behaviour as described, and negative scores represent
attitudes against the behaviour as described. Differences in range between predictor variables are acceptable for correlational
analysis, but if you want to compare the absolute values of predictor variables within or across studies, it would be necessary
to calculate the mean of the multiplied scores (in this case, +27/4 = +6.75, possible range -21 to +21).

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

6 MEASURING SUBJECTIVE NORMS
Key:

iii
ii
i

very time consuming (allow weeks)
quite time consuming (allow days)
not very time consuming (allow hours)

6.1 Direct measurement of subjective norm
i
6.1.1

Procedure

•

Direct measurement involves the use of questions referring to the opinions of important people
in general (See Box 6.1).

•

Use the first three items in the format presented in Box 6.1, and additional items if they seem
appropriate and if questionnaire length is not a problem.

•

Where the response format completes an otherwise incomplete sentence (e.g. I should not / I
should …), arrange the items so that the ends of the scales are a mix of positive and negative
endpoints (See Box 6.1). However, where an item is a complete sentence, and the responses
range from ‘Strongly disagree’ to ‘Strongly agree’, endpoints should not be mixed.

Box 6.1

Example: A patient presents with lower back pain. The target behaviour is referring the
patient for x-ray.
negative
endpoint

1. Most people who are important to me think that
I should 1

7 I should not

refer patients who have lower back pain for x-ray.

2. It is expected of me that I refer patients who have lower back pain for x-ray.
Strongly disagree 1

7 Strongly agree

3. I feel under social pressure to refer patients who have lower back pain for x-ray.
Strongly disagree 1

7 Strongly agree

4. People who are important to me want me to refer patients who have lower back pain for x-ray.
Strongly disagree 1

6.1.2
•

7 Strongly agree

Scoring
Recode the items that have negatively worded endpoints on the right, so that high scores then
consistently reflect greater social pressure to do the target behaviour

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

•

It is important that the subjective norm items have high internal consistency, i.e. that the scores
on these items correlate highly with each other. (You may decide to omit items from the scale to
improve internal consistency.)

•

Calculate the mean of the item scores to give an overall subjective norm score

6.2 Indirect measurement of subjective norm: measuring normative beliefs and motivation to
comply
6.2.1

Stages of Development

A. i ii

6.2.2

Conduct an elicitation study to elicit commonly held beliefs: Identify groups,
organisations and categories of individuals (‘reference groups’) who are likely to
apply social pressure with respect to the behaviour.
Construct questionnaire items to assess strength of normative beliefs with respect
to each reference group.
Construct questionnaire items to assess motivation to comply: Add items in
standard format for assessing motivation to comply with pressure from each
reference group.

Procedure
A Conduct an elicitation study to elicit commonly held beliefs

Box 6.2

•

Sample about 25 people from the same population from which you will select
respondents for the questionnaire study. (This would be the same 25 people as those
referred to in Section 5.2.2.)

•

Content analyse the responses into themes (normative beliefs) and label the sources of
social pressure extracted. At least two researchers should do this independently. List
these sources in order, from most frequently mentioned to least frequently mentioned.
Example: Your patient has Type 2 diabetes. The target behaviour is measuring the patient’s
blood pressure (BP). Please take a few minutes to list your thoughts about the following
questions:

Are there any individuals or groups who would approve of your measuring the patient’s BP?
Are there any individuals or groups who would disapprove of your measuring the patient’s BP?
Is there anything else you associate with measuring the patient’s blood pressure?

Construct questionnaire items to assess strength of normative beliefs
•

Select the reference groups (or individuals) most often listed and convert these into the
‘stems’ of normative belief items (see Box 6.3). Inclusion of 75% of the groups or
individuals listed should give adequate coverage of the sources of social pressure.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

•

Box 6.3

Items may reflect what important people think a person should do (injunctive norms) or what
important people actually do (descriptive norms). Box 6.3 illustrates the difference between
these types of items.

Imagine that the elicitation study has identified three sources of social pressure: patients with Type 2
diabetes; diabetologists; and other GPs.

Injunctive items (what important people think a person should do)
1. Patients with Type 2 diabetes think I
should not -3
-2
-1
0
+1
+2
measure
their
blood
pressure.
stem
2. Diabetologists would
disapprove -3

+3 should

-2
-1
0
+1
+2
+3 approve
of my measuring patients’ blood pressure.

Descriptive items (what important people actually do)
3. Other GPs
do not -3
-2
-1
0
+1
+2
+3 do
measure the blood pressure of their patients with diabetes.

Construct questionnaire items to assess motivation to comply
•

Convert each of the sources of social pressure into the form of a statement about the importance
of the various sources of social pressure. (See Box 6.4.) By answering the questions, the
participant indicates the strength of motivation to comply with each reference group or
individual.

Box 6.4
1. Patients’ approval of my practice is important to me
Not at all 1
2
3
4

7 Very much

2. What diabetologists think I should do matters to me
Not at all 1
2
3
4

7 Very much

3. Doing what other GPs do is important to me
Not at all 1
2
3

7 Very much

•

Pilot test these items by asking about five respondents to answer the questions and tell you
whether they have any difficulty answering them. If necessary, modify the wording of the
questions.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

6.2.3

Scoring

For each normative belief, the belief score on the should/should not or do/do not scale is multiplied by
the score relating to the not at all/very much scale (See Box 6.5)14. The resulting are summed products
across all the beliefs to create an overall subjective norm score:

N = (a x d) + (b x e) + (c x f)

Formula 6.1
Where

N = total subjective norm score
a, b and c are scores for each of the three normative beliefs
d, e and f are scores for motivation to comply relating to each source of social pressure

Using this method,
•
•

a positive (+) score means that, overall, the participant experiences social pressure to measure
the BP of patients for with diabetes.
a negative (-) score means that, overall, the participant experiences social pressure not to
measure the BP of patients for with diabetes.

Box 6.5 Example, scoring procedure
a

Patients with Type 2 diabetes
think I … measure their blood
pressure.

Diabetologists would ……. of my
measuring the BP of these
patients.
Other GPs ….. measure the BP
of these patients.

Should not

-3

-2

-1

Should

Disapprove

-3

-2

-1

+3 Approve

Do not

-3

-2

-1

Patients’ approval of my practice
is important to me.

Not at all

Very
much

What diabetologists think I should
do matters to me.

Not at all

Very
much

Doing what other GPs do is
important to me.

Not at all

Very
much

Imagine that a participant has responded by circling the numbers indicated in bold above.
The total normative belief score is calculated as
N = (+1 x 4) + (+3 x 1) + (+2 x 2)
= (+4) + (+3) + (+4)
= +11
The possible range of total scores is -63 to +63. THEREFORE,
THE NORMATIVE BELIEF SCORE OF THE PARTICIPANT REFLECTS FAIRLY WEAK
POSITIVE SOCIAL PRESSURE (i.e. TO MEASURE PATIENTS’ BLOOD PRESSURE).

Note that this form of scoring is somewhat controversial. Our reasons for recommending this approach are given in a
discussion paper (Appendix C).

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

7 MEASURING PERCEIVED BEHAVIOURAL CONTROL
Key:

iii
ii
i

very time consuming (allow weeks)
quite time consuming (allow days)
not very time consuming (allow hours)

7.1
i

Direct measurement of perceived behavioural control (PBC)

7.1.1

Procedure

•

Items should reflect people’s confidence that they are capable of performing the target
behaviour. This can be achieved assessing the person’s self-efficacy and their beliefs about the
controllability of the behaviour.

•

Self-efficacy is assessed by asking people to report
a) how difficult it is to perform the behaviour.
b) how confident they are that they could do it. (See Box 7.1)

•

Controllability is assessed by asking people to report
a) whether performing the behaviour is up to them. (See Box 7.1)
b) whether factors beyond their control determine their behaviour.

•

Remember that where the response format completes an otherwise incomplete sentence (e.g.
difficult / easy), arrange the items so that the ends of the scales are a mix of positive and
negative endpoints. (See Box 7.1.) However, where an item is a complete sentence, endpoints
should not be mixed.

Box 7.1
Self-efficacy
1. I am confident that I could refer my patients for x-ray if I wanted to
Strongly disagree 1

negative
endpoint

7 Strongly agree

2. For me to refer my patients for x-ray is
easy 1

7 difficult

Controllability
4. The decision to refer for x-ray is beyond my control.
Strongly disagree 1

7 Strongly agree

5. Whether I refer for x-ray or not is entirely up to me.
Strongly disagree 1

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

7.1.2

Scoring

•

Recode the items that have negative endpoints on the right, so that high scores then consistently
reflect a greater level of control over the target behaviour.

•

It is important that the subjective norm items have high internal consistency, i.e. that scores on
these items correlate highly with each other. (You may decide to omit items from the scale to
improve internal consistency.)

•

Calculate the mean of the item scores to give an overall perceived behavioural control score

7.2

Indirect measures of PBC: Measuring control beliefs and their perceived power to
influence behaviour

7.2.1

Stages of Development

A. i i i
B. i i
C. i

Conduct an elicitation study to elicit commonly held beliefs: Identify the content of
control beliefs which are shared by the target population about the behaviour.
Construct questionnaire items to assess the strength of these control beliefs
Construct questionnaire items to assess the power of these control factors to
influence the behaviour

7.2.2

Procedure

Conduct an elicitation study to elicit commonly held beliefs
•

Sample about 25 people from the same population from which you will select respondents for
the questionnaire study. (Again, this would be the same 25 people as those referred to in
Sections 5.2.2 and 6.2.2.)

•

Use open-ended questions. These are normally presented in one-to-one interviews, but could
also be in focus group or questionnaire form. Give participants a few minutes to list their
thoughts in response to structured questions (Box 7.2).

•

Content analyse the information into themes (control beliefs) and order and label the themes
extracted. At least two researchers should do this independently. List the themes in order from
most frequently mentioned to least frequently mentioned.

Box 7.2

Example: Your patient has Type 2 diabetes. The target behaviour is taking a
patient’s blood pressure. Please take a few minutes to list your thoughts about the
following questions:

What factors or circumstances enable you to measure the blood pressure of a patient with diabetes?
What factors or circumstances make it difficult or impossible for you to measure the blood pressure of
a patient with diabetes?
Are there any other issues that come to mind when you think about measuring the blood pressure of a
patient with diabetes?

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

B Construct questionnaire items to assess the strength of control beliefs
•

Select the beliefs most often listed and convert these into a set of statements. These statements
should reflect the beliefs which might make it difficult to perform (or not perform) the target
behaviour. (See Box 7.3 for question and response formats). Inclusion of 75% of all beliefs
listed should give adequate coverage of the belief ‘population’.

Box 7.3

Imagine that the elicitation study has identified a control factor to do with patients being
inappropriately dressed for BP measurement; another to do with feeling rushed when
measuring BP in the consultation; another about uncomfortable cuffs on BP machines.

1. Patients with diabetes come to the consultation inappropriately dressed to have their BP measured.
Unlikely 1
2
3
4
5
6
7 Likely
2. When I am measuring BP in the consultation I feel rushed.
Unlikely 1
2
3
4
5
6

7 Likely

3. The cuffs on the BP machines are uncomfortable for patients.
Unlikely 1
2
3
4
5
6

7 Likely

C Construct questionnaire items to assess the power of these factors to influence the behaviour
Convert each of the control belief statements into the form of an incomplete statement about whether
this makes it more or less likely that the person will do the target behaviour, or whether it makes the
behaviour easier or more difficult to do. (See Box 7.4.)
Box 7.4
1. When patients with diabetes come to the consultation inappropriately dressed to have their BP
measured, I am
less likely -3

-2

-1

+3 more likely

+3 much easier

to measure their BP.
2. Feeling rushed in the consultation makes it
much more difficult -3

-2

-1

to measure patients’ BP.
3. When the cuffs on the BP machine are uncomfortable for patients, I am
less likely -3

-2

-1

+3 more likely

to measure patients’ BP.

•

7.2.3

Pilot test these items by asking about five respondents to answer the questions and tell you
whether they have any difficulty answering them. If necessary, modify the wording.
Scoring

For each control belief, the belief score on the unlikely/likely scale is multiplied by the score relating to
the relevant item on the less likely/more likely scale or the much more difficult/much easier scale (See
Box 7.5). The resulting products are summed across all beliefs to create an overall perceived
behavioural control score:
23

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

PBC = (a x d) + (b x e) + (c x f)

Formula 7.1

Where

PBC = total perceived behavioural control score.
a, b and c are scores for each of three control beliefs.
d, e and f are scores for control belief power relating to each control belief.

Using this method,
•
•

a positive (+) score means that, overall, the participant feels in control of measuring patients’
blood pressure.
a negative (-) score means that, overall, the participant does not feel in control of measuring
patients’ blood pressure.

Box 7.5 Example, scoring procedure
a

Patients with diabetes come to the
consultation inappropriately dressed
to have their BP measured.

Unlikely

Likely

When I am measuring BP in the
consultation I feel rushed.

Unlikely

Likely

The cuffs on the BP machines are
uncomfortable for patients.

Unlikely

Likely

When patients with diabetes come to
the consultation inappropriately
dressed to have their BP measured,
I am ………….. to measure BP.

Less
likely

-3

-2

-1

More
likely

Feeling rushed in the consultation
makes it ……………. to measure
patient’s BP.

More
difficult

-3

-2

-1

Easier

When the cuffs on the BP machines
are uncomfortable for patients, I am
………….. to measure BP.

Less
likely

-3

-2

-1

More
likely

Imagine that the participant has responded by circling the numbers indicated in bolded italics above.
The total perceived behavioural control score is calculated as
PBC = (5 x -3) + (3 x -2) + (2 x -3)
= (-15) + (- 6) + (- 6)
= - 27

The possible range of total scores is -63 to +63. THEREFORE,
THE PBC SCORE OF THE PARTICIPANT REFLECTS A MODERATE LEVEL OF NEGATIVE
CONTROL, i.e. MEASURING PATIENTS’ BLOOD PRESSURE IS FAIRLY DIFFICULT.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

8 STEPS IN MANAGING A TPB SURVEY

8.1

Defining the population, clinical condition and behaviour of interest

Practical considerations and the more general research agenda will influence these decisions. For
example, in the context of implementation research, the project will be easier to complete and more
informative if
•
•
•
•
•
•

There is clear clinical evidence about the relevant issue(s)
compliance with the evidence is low or moderate
the clinical condition is not rare
among the population of interest there is variation in performing the behaviour to be
investigated
the behaviour involves a ‘yes/no’ treatment decision
contact details are easily available for the population of interest

After these definitions are agreed on by the project team, write a formal definition of the behaviour
under investigation. Recall the TACT principle, whereby the behaviour is defined with reference to its
Target, Action, Context and Time. You are then ready to select people from the population of interest
for the elicitation study.

8.2

Elicitation Study

8.2.1

Collecting the data

ONE elicitation study can be used to develop the indirect (belief-based) measures for all the predictor
constructs in the TPB model (attitude; subjective norm; and perceived behavioural control). This
involves:
•
•
•

8.2.2

selecting the sample for the elicitation study (ideally, 25 participants)
arranging a focus group or individual interview, or mailed questionnaire
asking for free-format responses to the set questions (see list of questions in Appendix A)

Analysing the data

Two researchers independently analyse the content of the responses by labelling the themes and listing
them in order of frequency for each of the following:
•
•
•

8.2.3

behavioural beliefs
sources of social pressure (reference individuals or groups)
control belief strength

Wording the questions

A questionnaire item is developed to assess each theme that has emerged from the analysis. At this
point, there should be a draft for questions to assess each of the following:
25

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

•
•
•

behavioural beliefs
sources of social pressure (reference individuals or groups)
control belief strength

Using similar wording, a second set of items is developed to assess each of the following:
•
•
•

outcome evaluation for each of the behavioural beliefs
motivation to comply with each reference group or individual
control belief power

8.3

Formatting and piloting the questionnaire

8.3.1

Number and content of questions

The first draft of the questionnaire includes:
•

A set of demographic questions which will provide information about the sample. Examples of
demographic questions are included in the questionnaire template (Appendix A).

•

Questions which provide direct measures of all three predictor variables.

•

The questions developed from the elicitation study, which are belief-based measures of the same
three predictor variables. It is important to measure all the constructs that are represented in the
model. All three predictors (attitude, subjective norm and perceived behavioural control) should
be included.

•

A set of questions to assess behavioural intention.

•

Each construct should be measured using a minimum of three items. In addition to the
demographic questions, this will result in a minimum of 12 items for intentions and direct
measures of the predictor variables and a further 18 items for belief-based measures. If you
decide to include the behavioural simulation method of measuring intentions, a further 10 items
(one for each scenario) would be used. Thus, the questionnaire should consist of a minimum of
40 carefully worded items, plus 8 demographic items.

•

Your elicitation study may reveal many more than three behavioural beliefs, normative beliefs
and control beliefs. You may feel that, in order to cover the breadth of the constructs of attitude,
subjective norm and perceived behavioural control, it is necessary to include a larger number of
items relating to indirect measures of the predictor variables. Including such extra items will
almost certainly improve the validity of the study but should be weighed against the issues of
questionnaire length and its consequences in terms of participant fatigue and response rates.

8.3.2

Ordering of questions in the document

Ajzen recommends that items be mixed up through out the document. That is, questions used to assess
attitudes should be interspersed with questions measure subjective norms and perceived behavioural
control.
26

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

8.3.3

Piloting the questionnaire and creating the second draft

Ask about five respondents to complete the questionnaire and comment on the items. (These may be
people who took part in the original elicitation study.) It would be useful if they could comment on the
following:
• Are any items ambiguous or difficult to answer?
• Does the questionnaire feel too repetitive?
• Does it feel too long?
• Does it feel too superficial?
• Are there any annoying features of the wording or formatting?
• Are there inconsistent responses that might indicate that changes in response endpoints are
problematic for respondents who complete the questionnaire quickly?
Cognitive interviewing techniques (Schwarz & Sudman, 1996) may be useful for pre-testing the
questionnaire items.
If responses from this small sample lead to minor changes in wording or formatting, a second draft of
the questionnaire can be written in consultation with the project team. If major changes are suggested, a
second elicitation study with a different sample would be required.

8.4

Brief forms of the questionnaire

Researchers may wish to construct a brief form of a TPB questionnaire. A subset of items may be
selected from the larger recommended questionnaire, but it is important that researchers are clear about
the purpose of the research, and understand which research questions can and cannot be answered by a
reduced data set. For example, the following list of broad research goals includes some ideas about
variables that would have to be measured to address each goal.
•

Predicting intentions

If the goal of the research is simply to do an analysis to predict variance in behavioural intentions, it
would be sufficient to measure intentions (3 generalised intention items) and the three predictor
variables using direct measures (3 items x 3 variables), resulting in a 12-item questionnaire.
•

Understanding influences

If the goal of the research is to identify the specific beliefs that contribute most to the three predictor
variables, both direct and indirect measures are needed. It is possible that a project may focus on
understanding one of the predictors, and this would permit a detailed exploration of the components of
that predictor (e.g. attitudes). However, there is the risk that other predictors may turn out to have a
stronger relationship with intentions.
•

Designing interventions

If the goal of the research is to assess the influence of each predictor with a view to designing an
intervention to modify the most powerful predictor, a brief questionnaire such as that described in Point
1 above would give some relevant information. Data about the specific beliefs that are most strongly
related to each predictor would be lacking, but could be assessed using a qualitative study of a subset of
participants.
27

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

•

Evaluating interventions

It is possible to assess one or some of the variables in the model using only the direct measures for
extremely brief versions of questionnaires in a before-after design, to determine the effect of an
intervention on any of the variables15. However, there is a risk associated with repeated use of the TPB
questionnaire, as it is possible that learning effects will contaminate responses. For this reason, it is
advisable to construct and pilot test parallel versions of a questionnaire (i.e. different items tapping the
same constructs), or else to use the same questionnaire within an RCT design.

8.5

Translating questionnaires into other languages

Translation involves not only linguistic issues but also theoretical issues and questions of
generalisability. The following broad principles are offered for researchers’ consideration:
•

The most important principle is that translation should be done by people who are native
speakers of the language to which the questionnaire is being translated (the ‘target language’).
This person should also be experienced in questionnaire design in the target language, and
ideally, also familiar with the TPB and ways in which TPB questionnaires are normally worded
in the target language. Note the words of the philosopher of science, Karl Popper:
‘Everybody who has done some translating, and who has thought about it, knows that there is no such
thing as a grammatically correct and also almost literal translation of any interesting text. Every good
translation is an interpretation of the original text; and I would even go so far as to say that every
good translation of a nontrivial text must be a theoretical reconstruction.’
(Popper, 1974/1992, p. 23)

•

The translated questionnaire is then subjected to a process of validation by using ‘backtranslation’ methods (translating back into the original language to establish equivalence with
the original version) and a pilot study (as described in 8.3.3 above).

•

Because it may not be valid to assume that commonly held beliefs about the target behaviour are
the same in the target language environment as in the original population of interest, it may be
appropriate to construct an entirely new questionnaire based on a new elicitation study using a
sample drawn from the new population. However, if the goal of the research involves a direct
comparison between countries with respect to specific beliefs and their influence on intentions,
it may be appropriate to translate directly from an English-language questionnaire. In that case,
it would be important to establish validity independently by conducting a qualitative study in the
target population. This study would investigate the commonly held beliefs relating to the
predictor variables and would assist in interpretation of findings from the comparative
quantitative study.

•

Sample questionnaires (brief form) about measuring blood pressure of patients with diabetes are
included in Appendix B in Dutch and Italian. These questionnaires have been subjected to the
back-translation and validation piloting processes described in the second dot point above.

Some researchers have been concerned that using a questionnaire in the control arm of a trial constitutes an intervention,
as asking the questions might influence behaviour. Our response to this concern is, ‘If only changing behaviour could be as
easy as that!’ This issue is discussed in greater detail in Appendix C.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

8.6

Main study: Practical issues

8.6.1

What sample size is needed?

Required sample size is determined by statistical power analysis. This requires the specification of the
study design and the expected effect size (Everitt, 1996). It is reasonable to assume at least a moderate
effect size (i.e. multiple R of around 0.3; Cohen, 1988) for TPB studies using a multiple regression
approach. Generally, a sample size of 80 would be acceptable. Note that response rates are often around
50%, so you need to send out 160 questionnaires to achieve this sample size unless you have reasons for
thinking that the response rate will be better than 50%.
Note that it is important to establish the representativeness of the sample, either by reporting a very high
response rate or by comparing the known characteristics of responders and non-responders.
8.6.2
•
•
•
•
•

Other aspects of survey methodology that need to be kept in mind
Approval is probably required from the relevant research ethics committee before the
commencement of the project. In the UK, compliance with research governance structures is
also required.
When the questionnaire is mailed to research participants, it should be accompanied by an
appropriate cover letter.
Reminder letters should be sent out to non-responders two weeks after the questionnaires are
mailed. A further reminder letter after a further two weeks will probably result in a further small
increase in the response rate.
It would be appropriate to close the data set after two months from the time of mailing the
questionnaire.
Readers may already be very familiar with these procedures, or may wish to consult the
following references for further details:

McColl E, et al. (2001). Design and use of questionnaires: A review of best practice applicable to
surveys of health service staff and patients. Health Technology Assessment Methodology, v. 5,
no. 31; Alton: Core Research on behalf of the NCCHTA.
Moser CA, Kalton G. (1993). Social methods in social investigation (2nd ed). Aldershot, Hants,
England: Dartmouth Publishing Co.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

DATA ENTRY AND ANALYSIS

Many readers may already be very familiar with the following processes, or may wish to consult the following
reference (or one similar) for further details:
Tabachnick BG, Fidell LS. (2001). Using Multivariate Statistics (4th Ed). Boston: Allyn and Bacon.

9.1 Setting up the data file
It is best to enter all the data into one SPSS file. Listing all the variables (questionnaire items) in the same order
in which they appear in the questionnaire makes data entry quick and undemanding. It is worth spending the time
to make clear and meaningful entries in the ‘variable label’ column. This will ensure that all output files include
the meanings and/or content of the items analysed, which saves a lot of time at the interpretation stage.
Do not attempt to sort variables or compute composite variables at this stage. The SPSS syntax files will do this
for you.

9.2 Data screening
Inspect the distributions of each variable, checking for data entry errors by noting whether all responses are in the
range represented by the response format. Highly skewed distributions may require the use of nonparametric tests
instead of a multiple regression approach to the analysis.

9.3 Main Analysis
•

Analysis using the direct measures of the predictor variables

Use the ‘recode’ command to recode any negatively worded responses. After this, conduct an item analysis on
the items relating to the direct measures, to establish internal consistency. If all internal consistency co-efficients
are acceptable (> 0.6 as a rough guide), it is appropriate to include all the items in the composite variables.
Use ‘compute’ commands to create the composite variables for the direct measures. Remember to define these
new variables clearly so that the variable labels will be included in the output files.
Using a multiple regression procedure, enter intention as the dependent variable, and the direct measures of
attitude, subjective norm and perceived behavioural control as the predictor variables.
•

Analyses using the indirect measures

Weight (multiply) each behavioural belief by the score for the relevant outcome evaluation to create a new
variable that represents the weighted score for each behavioural belief. Similarly, weight each normative belief
by the score for motivation to comply and each control belief by the score representing the influence of the
control belief. Then sum the weighted beliefs to create a composite score for attitude, subjective norm and
perceived behavioural control. It is a good idea to calculate a series of simple bivariate correlations between
direct and indirect measures of the same construct, to confirm the validity of the indirect measures. (Low
correlations would likely be a result of indirect measures that were poorly constructed or did not adequately
cover the breadth of the measured construct.)
Using a multiple regression procedure, enter directly-measured attitude scores as the dependent variable, and the
sum of the weighted behavioural beliefs as the predictor variables. Use a similar approach to predict directly
measured subjective norms and perceived behavioural control.
You may be interested in determining the specific beliefs that have the greatest influence on intentions. To do
this, dichotomise the intention variable either using a median split (i.e. low intenders versus high intenders) or by
classifying cases on a zero/greater than zero basis (i.e. non-intenders versus intenders). Use a series of t-tests or
discriminant analyses to identify the beliefs that discriminate between the two groups.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

10 REFERENCES AND SOURCES
Armitage CJ & Conner M. (2001). Efficacy of the Theory of Planned Behaviour: A meta-analytic review. British
Journal of Social Psychology, 40, 471-499.
Ajzen, I. (1985). From intentions to actions: A theory of planned behaviour. In J Kuhl, J Beckman (Eds), Action
control: From cognition to behaviour (pp. 11-39). New York: Springer.
Ajzen, I. (1988). Attitudes, personality and behaviour. Milton Keynes; OUP.
Ajzen I. (1991). The theory of planned behaviour. Organizational Behaviour and Human Decision Processes, 50,
179-211.
Ajzen I. (2003). Website: http://www-unix.oit.umass.edu/%7Eaizen/
Cohen J. (1988). Statistical power analysis (2nd ed.). Hillsdale, NJ: Erlbaum.
Conner M, Sparks P. (1995). The Theory of Planned Behaviour and health behaviours. In M Conner, P Norman
(Eds), Predicting health behaviour pp. 121-162. Buckingham: OUP.
Everitt, B. (1996). The Cambridge dictionary of statistics in the medical sciences. Cambridge, England:
Cambridge University Press.
Fishbein M. (1967). Attitude and the prediction of behavior. In M Fishbein (Ed.), Readings in attitude theory and
measurement. New York: Wiley.
Gagné C, Godin G. (1999). Les théories socials cognitives: Guide pour la mesure des variables développement de
questionnaire. Groupe de recherché sure les aspects psychosociaux de la santé, École des sciences
infirmieres, Université Laval.
Godin G, Kok G. (1996). The Theory of Planned behaviour: A review of its applications to health-related
behaviours. American Journal of Health Promotion, 11(2), 87-98.
Jones TV, Gerrity MS, Earp J. (1990). Written case simulations: Do they predict physicians’ behaviour? Journal
of Clinical Epidemiology, 43, 805-815.
McColl E, et al. (2001). Design and use of questionnaires: A review of best practice applicable to surveys of
health service staff and patients. Health Technology Assessment Methodology, v. 5, no. 31; Alton: Core
Research on behalf of the NCCHTA.
Moser CA, Kalton G. (1993). Social methods in social investigation (2nd ed). Aldershot, Hants, England:
Dartmouth Publishing Co.
Popper K. (1974). Unended quest: An intellectual autobiography (updated edition 1992). London: Routledge.
Seale C. (1998). (ed.) Researching society and culture. London : Sage.
Tabachnick BG, Fidell LS. (2001). Using Multivariate Statistics (4th Ed). Boston: Allyn and Bacon.
Valois P, Godin G. (1991). The importance of selecting appropriate adjective pairs for measuring attitude based
on the semantic differential method. Quality and Quantity, 25, 57-68.
Walker AE, Grimshaw JM, Armstrong EM. (2001). Salient beliefs and intentions to prescribe antibiotics for
patients with a sore throat. British Journal of Health Psychology, 6, 347-360.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

11 GLOSSARY16
Attitude

A psychological tendency that is expressed by evaluating a
particular behaviour with some degree of favour or disfavour

Behaviour

An action that is carried out at a specified time and is
described in terms of the action itself, its target and the context

Behavioural beliefs

The perceived consequences of an action

Cognitive
interviewing

A technique of pre-testing questionnaire items, in which
respondents are encouraged to share their thoughts with the
researcher about items either concurrently (as they answer the
items) or retrospectively

Content analysis

Qualitative analysis of verbal data to discover the underlying
topics or themes. References to these themes are often then
counted to determine the most frequently mentioned themes

Control beliefs

Beliefs about the likelihood that one possesses the resources
and opportunities thought necessary to execute the behaviour

Elicitation study

A qualitative investigation of a subset of a population under
investigation, to discover the salient behavioural, normative
and control beliefs about the behaviour

Endpoints

Verbal labels that are written at each end of a row of numbers
to indicate the meanings of the most extreme numbers

Intention

A person’s motivation in the senses of his or
her conscious plan to exert effort to carry out a behaviour

Internal consistency

See ‘reliability’

Motivation to comply

The extent to which a person feels inclined to match his or her
behaviour to various sources of social pressure

Multiple regression

A quantitative analytic procedure that either simultaneously or
cumulatively assesses correlations between a number of
independent variables and one dependent variable

Norms

Normative beliefs

-descriptive

Perceptions about what important people actually do

-injunctive

Perceptions about what important people think a person
should do

-subjective

Perceived social pressure to perform a behaviour
Perceptions of significant others’ preferences about whether
one should perform a behaviour

Many of these definitions are derived from the seminal text by Eagly and Chaiken (1993). For expanded explanations of these
definitions, this book is recommended. Also, readers are advised to refer to Everitt (1996) for other views.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

Outcome evaluation

Evaluation of the perceived consequences of an action

Perceived
behavioural control

Perceptions about how easy or difficult it is to perform the
behaviour

Reliability

A property of a measuring instrument, indicating the extent to
which it yields consistent results over repeated observations
-internal
consistency

A statistic for assessing the equivalence of different items in a
scale. It is appropriate for measuring the reliability of a scale
composed of multiple items, if it is valid to assume that the
items are parallel measures of the same attitude content
domain

-test-retest

A statistic for assessing the stability of a scale over time
(although this is subject to contamination by changes in ‘true
scores’ over time)

Self-efficacy

The conviction that one can successfully execute a given
behaviour

TACT principle

The description of behaviour in terms of its target, the action
itself, the context in which it is performed, and when it is
performed

Validity

A property of measuring instruments or of responses,
indicating the extent to which they measure what they are
supposed to measure

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

12 APPENDIX A: QUESTIONNAIRE DEVELOPMENT MATERIALS
12.1

Example, Elicitation Study Materials

The following may be used as a structured interview schedule for individual or focus group interviews,
or it may be converted to a written questionnaire with a free responses format. Words in [square
brackets] should be replaced in your own study, to reflect the population and the behaviour you decide
to investigate.

We are conducting a study of [GPs] in [North Tyneside]. We are interested in the reasons why [GPs] do
or do not [measure the blood pressure] of [their patients with Type 2 diabetes]. We would appreciate
your responses to some questions about this. There are no right or wrong answers. Please tell us what
you really think.
*********************************************************************************
Please take a few minutes to list your thoughts about the following questions.
When [patients with Type 2 diabetes] consult their [GP],

•

What do you believe are the advantages of [measuring the patient’s blood pressure during
a consultation]?

•

What do you believe are the disadvantages of [measuring the patient’s blood pressure
during a consultation]?

•

Is there anything else you associate with your own views about [measuring the patient’s
blood pressure during a consultation]?

•

Are there any individual or groups who would approve of your [measuring the patient’s BP
during a consultation]?

•

Are there any individual or groups who would disapprove of your [measuring the patient’s
BP during a consultation]?

•

Is there anything else you associate with other people’s views about [measuring the
patient’s blood pressure during a consultation]?

•

What factors or circumstances would enable you to [measure the blood pressure of a patient
with diabetes during a consultation]?

•

What factors or circumstances would make it difficult or impossible for you to [measure the
blood pressure of a patient with diabetes during a consultation]?

•

Are there any other issues that come to mind when you think about [measuring the blood
pressure of a patient with diabetes during a consultation]?

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

12.2

Example, Questionnaire

Note that this questionnaire has TWO sections. Section 1 asks about demographic information; Section
2 measures the predictor variables and intentions.

SECTION 1
About your BACKGROUND
Years

A How long have you been qualified?
How many sessions (1/2 days) do you work per week?
B How many GPs work at your practice?
C How many practice nurses work at your practice?
D Are you Male

Or Female

E Are you a GP

Or a Practice Nurse

If you are a GP please go to Question G. If you are a Practice Nurse please answer question F
F If you are a nurse do you provide care for people
with Type 2 diabetes?
Yes
No
G What is your approximate practice list size?

SECTION 2
Each question in this section refers to MEASURING THE
BLOOD PRESSURE (BP) of your patients with Type 2 diabetes
1

If I measure their BP, I will feel
that I am doing something
positive for the patient
It causes a lot of worry and
concern for a diabetes patient if
they are found to have high BP

Unlikely

Likely

Unlikely

If I measure BP, I will detect any
problems at an early stage.

Unlikely

Likely

If I measure BP, I’ve got to see
patients more often

Unlikely

Likely

The blood pressure machine I use
is not very accurate

Unlikely

Likely

When I am measuring BP in the
consultation I feel rushed

Unlikely

Likely

The cuffs on BP machines are
uncomfortable for patients

Unlikely

Likely

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

Patients come to the consultation
inappropriately dressed to have
their BP measured

Unlikely

Likely

Seeing patients more often is

Extremely

-3

-2

-1

Extremely
desirable

-3

-2

-1

Extremely

undesirable

Worry and concern experienced
by patients if they are found to
have high BP is
Doing something positive for the
patient is

Extremely

For these patients, detecting
problems at an early stage is

Extremely

Diabetologists think I

should not

undesirable

Extremely

desirable

-3

-2

-1

undesirable

Extremely
desirable

-3

-2

-1

undesirable

Extremely
desirable

-3

-2

-1

should

measure the BP of patients with diabetes.
14

Patients with Type 2 diabetes
would

disapprove

-3

-2

-1

approve

of my measuring their BP.
15

Other GPs

do not

-3

-2

-1

measure the BP of their patients with diabetes.
16

The government would

disapprove

-3

-2

-1

approve

of my measuring the BP of patients with diabetes.
20

Overall I think that measuring
these patients’ BP is:

Harmful

Beneficial

Pleasant

Unpleasant

The wrong
thing to do

The right
thing to do

Good
practice

Bad practice

Doing what other GPs do is
important to me

Not at all

Extremely

Doing what diabetologists think I
should do is important to me

Not at all

Extremely

The government’s approval of my
clinical practice is important to me

Not at all

Extremely

The approval of my patients with
Type 2 diabetes is important to
me

Not at all

Extremely

I am

Less likely

-3

-2

-1

More likely

21
22
23
24

to measure patients’ BP if the machine that I use is accurate.

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

Less likely

I am

-3

-2

-1

More likely

to measure patients’ BP if the cuffs on the machines are uncomfortable.
30

Less likely

I am

-3

-2

-1

More likely

to measure patients’ BP if I feel rushed in the consultation.
31

Less likely

I am

-3

-2

-1

More likely

to measure patients’ BP if they come to the consultation
inappropriately dressed to have their BP measured.
32

People who are important to me
think that I should NOT measure
the BP of patients with Type 2
diabetes

Strongly
disagree

Strongly
agree

I expect to measure the BP of my
patients with diabetes in each
consultation

Strongly
disagree

Strongly
agree

Strongly
disagree

Strongly
agree

I am confident that I can measure
the BP of these patients in the
consultation if I want to.

Strongly
disagree

Strongly
agree

Whether I measure the BP of
these patients in the consultation
is entirely up to me

Strongly
disagree

Strongly
agree

For me to measure the BP of
these patients in the consultation
is

easy

difficult

I want to measure the BP of my
patients with diabetes in each
consultation

Strongly
disagree

Strongly
agree

Strongly
disagree

Strongly
agree

Strongly
disagree

Strongly
agree

I feel under social pressure to
measure the BP of these patients

It is expected of me that I
measure the BP of these patients
I intend to measure the BP of my
patients with diabetes in each
consultation

Out of the next 10 patients you see with
a diagnosis of Type 2 diabetes, for how
many would expect to measure BP?

THANK YOU FOR YOUR PARTICIPATION
38

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

12.3

Example, scoring key for questionnaire

After constructing your questionnaire, it is important to construct a scoring key such as the one below.
This guides various stages of data analysis.

Response
format
Question
Numbers

Items
requiring
reverse
scoring

Items
requiring
internal
consistency
analysis

Items
requiring
multiplication
Construct measured

1 to 4

1 to 7

1 x 11; 2 x 10;

Behavioural Beliefs

9 to 12

-3 to +3

3 x 12; 4 x 9

Outcome Evaluations

13 to 16

-3 to +3

Normative Beliefs

24 to 27

1 to 7

13 x 25;
14 x 27;
15 x 24;
16 x 26

5 to 8

1 to 7

5 x 28; 6 x 30;

Control Belief Strength

28 to 31

-3 to +3

7 x 29; 8 x 31

Control Belief Power

20 to 23

1 to 7

Motivation to Comply

21 and 23

20 to 23
(after
recoding)

32 (after
recoding),
34 and 39

Control, direct measure

Generalised intention

Attitudes, direct measure
Subjective Norms, direct

32, 34, 39

1 to 7

measure
Perceived Behavioural

35 to 37

1 to 7

35, 36 and
37 (after
recoding)

33, 38, 40

1 to 7

33, 38, 40

0 to 10

Intention statement

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

13 APPENDIX B: EXAMPLES OF BRIEF QUESTIONNAIRES IN LANGUAGES OTHER
THAN ENGLISH
13.1

Dutch language17

Behaviour: Measuring the blood pressure of patients with diabetes
Elke vraag in deze sectie heeft betrekking op de behandeling van uw patiënten met
Type 2 diabetes. Wilt u voor elke vraag alstublieft telkens één nummer omcirkelen dat
het meeste met uw mening overeenkomt. De nummer slopen van 1 (geheel mee oneens)
tot 7 (geheel mee eens).
1.
2.
3.
4.
5.

Mensen die belangrijk voor mij zijn vinden dat ik de bloeddruk
van mijn diabetespatiënten moet meten
Mijn professionele organisatie vindt dat ik de bloeddruk van mijn
diabetespatiënten moet meten
Het meten van de bloeddruk van mijn diabetespatiënten is
moeilijk
Ik ben van plan om de bloeddruk te meten van al mijn
diabetespatiënten tijdens hun volgende consult
Ik verwacht dat ik de bloeddruk van al mijn diabetespatiënten ga
meten tijdens hun volgende consult
In het algemeen is volgens
mij het meten van de
bloeddruk van mijn
diabetespatiënten:

Geheel
mee oneens
1 2 3

Geheel
mee eens
5 6 7

Schadelijk

gezondheids
bevorderend

Prettig

Niet prettig

Verkeerd
om te doen

Juist om te
doen

Goed
handelen

Slecht
handelen

Geheel
mee oneens
1 2 3
10. Ik wil de bloeddruk meten van al mijn diabetespatiënten tijdens

hun volgende consult
11. Andere huisartsen meten de bloeddruk van hun diabetespatiënten
niet
12. Ik bepaal geheel zelf of ik de bloeddruk van mijn
diabetespatiënten meet
13. In het algemeen voel ik mij niet in staatom de bloeddruk van
mijn diabetespatiënten temeten

Geheel
mee eens
5 6 7

We are extremely grateful to Wendy Hardeman, Marije Bosch and Dr Trudy van der Weijden, who provided translations,
back-translations and comments on this questionnaire

Constructing Theory of Planned Behaviour Questionnaires: Manual for Researchers

13.2

Italian language18

Behaviour: Measuring the blood pressure of patients with diabetes
Ciascuna domanda di questa sezione si riferisce al trattamento dei Suoi pazienti con un
diabete di tipo 2. Per ciascua affermazione, la preghiamo di cerchiare il numero
indicante il suo livello di accordo, da 1 (completamente in disaccordo) a 7
(completamente d’accordo) con i vari passi intermedi.

1.
2.
3.
4.
5.

Completamente
in disaccordo
1 2 3
Le persone che sono per me importanti professionalmente pensano
che dovrei misurare la pressione del sangue dei miei pazienti con
La mia societa’ scientifica/professionale pensa che dovrei misurare 1 2 3
la pressione del sangue dei miei pazienti con diabete
1 2 3
Misurare la pressione del sangue dei miei pazienti con diabete e’
difficile
Intendo misurare la pressione del sangue di tutti i miei pazienti con 1 2 3
diabete alla loro prossima visita
1 2 3
Mi aspetto di misurare la pressione del sangue di tutti i miei
pazienti con diabete alla loro prossima visita

Tutto considerato, penso che
misurare la pressione del
sangue dei miei pazienti con
diabete sia:

Completamente
d’accordo
4 5 6 7
4

Dannoso

Benefico

Gradevole

Sgradevole

La cosa
sbagliata
da fare

La cosa
giusta da
fare

7.
8.

9.
Una
buona
prassi

10.
11.
12.
13.

Completamente
in disaccordo
1 2
Voglio misurare la pressione del sangue di tutti i miei pazienti
con diabete alla loro prossima visita
Altri medici di famiglia non misurano la pressione del sangue dei 1 2
loro pazienti con diabete
1 2
Misurare la pressione del sangue dei miei pazienti con diabete
dipende interamente da me
1 2
Tutto considerato, non credo di poter misurare la pressione del
sangue dei miei pazienti con diabete

Una cattiva
prassi

Completamente
d’accordo
4 5 6 7

We are extremely grateful to Dr Marco Perugini and Dr Luciana Ballini, who provided translations, back-translations and
comments on this questionnaire

File Type	application/pdf
File Title	Microsoft Word - TPB Manual FINAL May2004.DOC
Author	njl27
File Modified	2012-08-18
File Created	2004-05-05