Enviropreneurial orientation in SME supply chains: construct measurement development
| Pages | 128-150 |
| Published date | 08 May 2017 |
| DOI | https://doi.org/10.1108/WJEMSD-08-2016-0036 |
| Date | 08 May 2017 |
| Author | Sheila Namagembe,Suzanne Ryan,Ramaswami Sridham |
| Subject Matter | Strategy,Business ethics,Sustainability |
Enviropreneurial orientation in
SME supply chains: construct
measurement development
Sheila Namagembe, Suzanne Ryan and Ramaswami Sridham
Newcastle Business School, University of Newcastle, Callaghan, Australia
Abstract
Purpose –The purpose of this paper is to propose a new model that is inclusive andpractical because of the
deficiencies in models for construct measurement. Further, the authors demonstrate the value of the proposed
model by describing its application to the development and validation of a multi-dimensional construct,
enviropreneurial orientation. Although used in the literature, enviropreneurial orientation had not been
developed nor tested as a construct. The paper provides detailed explanation of development and validation
processes exemplified by experiences of research into the factors that motivate individual enviropreneurial
orientation among owner/managers in supply chains.
Design/methodology/approach –The authors review models of construct development and propose an
eight step model to overcome the deficiencies in the existing models. The eight steps are: defining the
construct; identifying its dimensions; generating measurement items for each dimension; pretesting
the measurement items; collecting data; constructing scales; analysing reliability; and evaluating the
relationships. Each step is explained through examples based on the authors’experience in using the model to
develop the new construct –enviropreneurial orientation.
Findings –All correlation results were positive and significant as were the multiple regression results with
one exception, competitive aggressiveness.
Originality/value –The authors provide a practical model to guide new construct measurement
development which can be used by researchers and research students for multi-dimensional constructs,
especially they are constrained by time and financial resources. By using an example to demonstrate the
applicability of the model, the authors go beyond the usual description of construct development models to
make the proposed model more comprehensible and thus useful.
Keywords Sustainability, Entrepreneurship, SMEs, Supply chain
Paper type Research paper
1. Introduction
The primary goal of measurement development is to create dimensions that measure the
underlying construct. Measurement development may involve revising dimensions for an
old construct or generating dimensions for a new construct (Churchill, 1979; DeVellis,
1991). Development of measures for new constructs, especially multi-dimensional
constructs, can be problematic because the researcher does not have existing measures to
build upon. Although measurement development for constructs is a common process in
quantitative research, a well-established framework to guide researchers through the
various stages of measurement development is lacking, so the efforts to develop new
constructs are often fragmented and incomplete (Hinkin, 1995). Thus the aim of our paper
is to propose a framework for measuring new constructs and demonstrate how it was
applied to the development of a new multi-dimensional construct, enviropreneurial
orientation in supply chains. Although the need for the development of the measures for
enviropreneurial orientation was recommended 20 years ago by Varadarajan (1992), no
such measures have been developed. We proceed with an evaluation of existing
measurement models before suggesting an eight step model based primarily on the
strengths of similar models proposed by Bollen (1989), Schwab (1980) and DeVellis (1991,
2003) . This is followed by an example of the model’s application to the development of
enviropreneurial orientation. Our contribution lies in providing a comprehensive but
practical model to guide researchers in the definition and measurement development of
World Journal of
Entrepreneurship, Management
and Sustainable Development
Vol. 13 No. 2, 2017
pp. 128-150
© Emerald PublishingLimited
2042-5961
DOI 10.1108/WJEMSD-08-2016-0036
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/2042-5961.htm
128
WJEMSD
13,2
new constructs. Because our example construct, enviropreneurial orientation, is in a
supply chain context, the paper has particular relevance to supply chain management
literature and researchers.
2. Models of construct measurement development
Measurement models guide researchers in the measurement development process for
new constructs. Measurement “consists of rules for assigning symbols to objects so as to
(1) represent quantities of attributes numerically (scaling) or (2) define whether the objects
fall in the same or different categories with respect to a given attribute (classification)”
(Nunnally and Bernstein, 1994, p. 3). Five commonly used measurement models are:
Churchhill’s (1979) eight stage model; Schwab’s (1980) three stage model; Bollen’s (1989) four
stage model; DeVellis’s (1991, 2003) eight stage model and Mackenzie et al. (2011) ten stage
model. We evaluate each of these models and propose a combined model primarily based on
Bollen (1989); Schwab (1980) and Devillis (2003).
First, Churchill’s (1979) measurement model involves eight stages: specifying the domain
of the construct; generating the sample of items; collecting (pilot) data; purifying through
conducting reliability and factor analysis; collecting data; assessing reliability; assessing
validity; and lastly developing norms. This model is commonly used in marketing research
(see Mackenzie et al., 2011; Churchill, 1979). It provides an ordered set of activities to guide
researchers in developing and evaluating measurement scales for multi-dimensional
constructs (Mackenzie et al., 2011). The model has two weaknesses. It fails to provide clarity
on sample size required to test constructs (Flynn and Pearcy, 2001) and the recommended
use of internal consistency to purify a measure without focus on the respondents may result
in deletion of important items for a construct (Smith, 1999).
Second, Schwab’s (1980) three stage model includes: item generation; scale construction;
and scale evaluation. While Schwab’s (1980) stages are common in organisational research
(see Chen and Chen, 2012; Hinkin, 1995; Mohamad et al., 2014), multiple tasks are required in
a single stage (Hensley, 1999; Hinkin, 1995). For example, stage one involves developing a
definition of the construct, generating dimensions and then measurement items for the
dimensions. The model does not specify whether this applies to first order reflective
constructs or second order formative constructs or both. However, the measurement
model is simple, suitable for data collection at single a point in time and can be used for
either a deductive or an inductive approach during construct measurement development
(see Hinkin, 1995).
Third, Bollen’s (1989) model has four stages: developing a definition for the concept;
identifying dimensions for a concept; developing measurement items for the dimensions;
and developing a measurement model that involves specification of relationships between
measurement items and the construct. Although Bollen’s (1989) model overcomes the
problem of reducing multiple activities into one stage, the model concludes with the need
to test for relationships but without specific mention of validity or reliability measures
(see Hu and Bentler, 1999; Lin, 2013).
Fourth, DeVellis (1991, 2003) suggests eight stages: determining what is to be measured;
generating the item pool; determining the format for measurement; reviewing initial items
with a panel of experts; considering inclusion of validation items; administering items to an
administrative sample; evaluating the items; and optimising scale length. Whereas
Devellis’s model is one of the commonly used models, focus is placed on reflective constructs
and fails to provide guidelines for testing the temporal stability of a measure (Dahly and
Adair, 2007). Nevertheless, the model may be applied under circumstances where a
researcher is constrained by time.
The final model, Mackenzie et al. (2011), involves ten stages: developing a conceptual
definition of the construct; generating items to rep resent the construct; assessing the content
129
SME supply
chains
To continue reading
Request your trial