Skip navigation

Category Archives: Construct Evaluation

Article by Barki, Henri ( 2008 ) in The DATA BASE for Advances in Information Systems, 39:3..


When I first read the title and the abstract of this paper, I wonder what is the meaning of ‘thar’. I tried to find it in dictionary. I couldn’t find any. I typed the word in MSWord and searched for thesaurus, it revealed nothing. I surfed the internet and searched the word over Google, I ended up suffocated with ocean of essays. Finally I searched the word over the Wikipedia encyclopedia, this time I got this article. I read it, I found it interesting.


Later when I read the introduction of the paper, I found that the author has already explained the meaning in the very beginning of the paper. Moral of the story, don’t judge a paper by its title and abstract!!


This paper describes 4 different systematic approaches (but not exclusively independent of each other) in conceptualizing (or developing) constructs[1]:


(1)  Make a Clear Definition

Once a concept has been identified to be conceptualized as a construct, a clear and explicit definition has to be established. For example, to conceptualize “project management” as a construct, a clear definition that is suitable to the research context has to be established. Although previous studies have identified core issues/facets of “project management” as ‘(i) project planning and (ii) monitoring’, ‘(iii) project coordination’ and ‘(iv) project control’, all of the four facets may not necessarily move in the same direction. It may (or may not) vary in different contexts. Therefore, to come out with a clear definition, researcher has to analyze the appropriateness (based on the research context) and decide either to ensemble or separate the core facets as the definition of the construct.


Another aspect to be considered in this approach is the appropriateness to separately define notions that belong to different meta-categories, for example to clearly distinguish between behavioral (i.e., activities), cognitive (i.e., beliefs), and attitudinal (i.e., feelings) constructs by defining them as different constructs. This step is very useful in explaining time-dependent relationships that frequently exist between the constructs of different meta-categories.


(2)  Specify Construct’s Dimensions and Their Relationships

Another way to conceptualize constructs is by identifying the dimensions (determinants) of a construct and the relationships that exist between them. This can be done by studying the literature such as findings from previous related studies or other field’s theories. The researcher has to analyze and decide whether or not all construct dimensions must be simultaneously present for the construct to exist.


(3)  Explore How Construct Applies to Alternative Context

The third approach in conceptualizing a construct is by analyzing further on how a given construct can apply in different technological, organizational, or individual contexts. Examples given by the author are constructs in a research model (such as TAM, TTF) where they are examined in a different context (such as hedonic versus instrumental contexts or between different tasks in e-commerce contexts).


(4)  Expand the Conceptualization of Construct

In the existing research models, constructs are ‘narrowly defined’ (to suit with particular context being studied). Hitherto, one can conceptualize better construct by expanding the conceptualization of the existing constructs to the extent that it would better explain and understand multifaceted and complex realities.


One very interesting point of discussion in this paper is the issue of construct evaluation. The author argues that the assessment of construct’s quality should not be focusing on the ‘truthfulness’[2] of the construct’s conceptualization, but should rather emphasized on the construct’s efficacy in predicting or explaining interesting or important phenomena. It is where, according to the author, the constructs contribute to the existing knowledge and theory.


[Acknowledgement: I would like to thank Assoc. Prof. Dr. Nor Laila Md Noor for giving me this article.]

[1] Definition of construct in Wikipedia: Construct is ‘an explanatory variable which is not directly observable. For example, the concepts of intelligence and motivation are used to explain phenomena in psychology, but neither is directly observable.


[2] Furthermore, there is no such thing as ‘true’ nature of a construct – it has infinite ways to conceptualize it!