Remember that when you chose your route (i.e., Route A: Duplication, Route B: Generalisation or Route C: Extension), you also chose a specific approach within that route (e.g., a population, context/setting, treatment or time-based generalisation within Route B: Generalisation). Now that you have a much better understanding of (a) the broader literature, (b) the main theoretical components of your main journal article, and (c) the support for and criticisms of the main journal article, you need to explain and justify the route you have chosen, and the approach within that route.
Common justifications for replication-based dissertations include reasons related to (a) improving the internal validity of a study, (b) testing the generalizability of a study, (c) building the construct validity and reliability of a measurement procedure, (d) testing whether different methods lead to the same results, and whether existing measures are sufficient, and (e) assessing the data analysis techniques used. However, from the understanding you have gained about the literature and your main journal article, you need to explain and justify the route you have chosen, and the approach within that route, in as much detail as possible. To start this process, ask yourself:
To what extent do the theoretical components that I have identified in my main journal article differ from those inapproach my chosen route, and the I have adopted within that route?
What do I need to look for in the literature in order to address these differences in theoretical components?
Now consider your route and approach:
If you are pursuing Route A: Duplication, there should be very little variation in the theoretical components between the main journal article and your route and approach since the essence of duplication is to replicate the research set out in the main journal article as closely as possible. Instead, the critical evaluation of the main journal article in the previous step (i.e., STEP TWO) should form the basis of your explanation and justification for adopting Route A: Duplication. Therefore, when you come to write up your dissertation, as well as set the research strategy for your dissertation (i.e., in STAGE SIX), you are likely to be able to rely on the theoretical components already set out in your main journal article. If you pursued Route A: Duplication, you may want to jump forward to STEP FOUR: Set the theoretical model for your dissertation.
For Route B: Generalisation, the majority of the theoretical components in the main journal article should be the same as those used in your route because generalisations only involve making changes to specific theoretical components based on the approach you have adopted.
In some cases, the type of generalisation you are making will only involve small changes to the main journal article. For example, if we were simply making population-based generalisations, and the new population was very similar to the population in the main journal article, few changes would be required (e.g., if we took a main journal article that examined the relationship between study time and student achievement amongst undergraduate students, and examined whether this relationship could be generalised to postgraduate students).
However, if the two populations were very different (i.e., the population from the main journal article and the new population), you would need to look to the literature to explain (a) why such differences were important and (b) how they could affect your dissertation. The same could be said if the approach to generalisation involved a particularly different treatment or context/setting between the main journal article and your dissertation (i.e., in the case of a treatment or context/setting-based generalisation). For example, imagine that the main journal article examined the relationship between ethical sales behaviour and customer loyalty on financial services customers, but your dissertation was interested in this same relationship, but amongst consumers on online auction sites. This would involve a population-based and setting/context-based generalisation. The implications of these different populations and settings/contexts on the relationship being investigated (i.e., between ethical sales behaviour and customer loyalty) could be very significant because (a) the types of customer are different, (b) there are two different industries involved - financial services and online auction sites - and (c) the contexts/settings are different - mainly face-to-face sales in financial services, with relationships between seller and buyer developed over the long-term, compared with an online selling environment without face-to-face contact, where there may be more short-term relationships between buyers and sellers - all of which could affect the theoretical components you need to take into account.
When it comes to writing up your Literature Review chapter (usually Chapter Two: Literature Review), you will need to be able to explain (a) what the differences in your population, context/setting, treatment or time are when compared with the main journal article, (b) why they are different (i.e., the differences in their characteristics, and perhaps any theoretical reasons for such differences), and (c) how this will affect the theoretical components that you will build on when it comes to setting the research strategy for your dissertation (i.e., in STAGE SIX). However, at this stage, the most important reason for looking to the literature to understand the what, why and how of such differences is because of the impact that these differences having when it comes to STAGE SIX: Setting your research strategy. For example, differences in the population or context/setting between your dissertation and the main journal article will affect the sampling strategy that you adopt, and may even affect the choice of research methods to collect your quantitative data, while differences in the treatment between your dissertation and the main journal article could affect everything from your choice of research design, through to your sampling strategy, research methods and data analysis techniques.
To identify the differences in theoretical components between your dissertation and the main journal article, we would recommend that you start by identifying the theoretical components that speak directly to the approach that you have adopted:
A population or context/setting-based approach to generalisation
When adopting a population or context/setting-based generalisation, start by focusing on the differences in the characteristics of the population or context/setting between the main journal article and your dissertation.
For example, imagine that the main journal article you are interested in examined the relationship between ethical sales behaviour and customer loyalty on financial services customers, and your dissertation was interested in this same relationship, but in a different population and context/setting, namely consumers on online auction sites. What is it about financial services customers that makes them different from consumers on online auction sites (e.g., amount of money such consumers spend on products, the types of products that are bought, the tangibility of the products, the demographics of the customers, etc.)? Similarly, what differences are there between financial services salespeople and salespeople on online auction sites (e.g., the proportion of professional versus private/amateur salespeople, the full-time versus part-time nature of the salespeople, the level of product knowledge, the length of the use of such products, the relationship between buyer and seller, etc.)? In what ways are the contexts/settings of financial services and online auctions different (e.g., type of customer contact - face-to-face versus distance/online only - level of risk in products being sold, relative information asymmetry of products, etc.)?
These differences in the characteristics of the population or context/setting between the main journal article and your dissertation will not only (a) affect the sampling strategy that you adopt, and may even affect the choice of research methods to collect your quantitative data, but (b) should also help to explain the differences between the results obtained in the main journal article and your dissertation. This is something we discuss at more length in STAGE SIX: Setting your research strategy and Chapter Four: Results. At this point, you need to search the literature for books or journal articles that help to explain the characteristics of the populations or contexts/settings that you are interested in.
A treatment-based approach to generalisation
When adopting a treatment-based generalisation, start by focusing on the differences in the characteristics of the treatment between the main journal article and your dissertation.
For example, imagine that the main journal article you are interested in examined the relationship between lecture length and knowledge retention, and your dissertation was interested in the same relationship, and the same population and context/setting. However, you were interested in whether the treatment used in the main journal article could be generalised. The treatment used in the main journal article involved testing the effect of just two lecture lengths (i.e., 1 hour long lecture and a 2 hour long lecture), and the findings showed that knowledge retention dropped significantly in the 2 hour lecture compared with the 1 hour lecture. However, the course schedule demands that professors teach longer lectures, where possible. You want to examine whether different lecture lengths, or lectures that involve a break rather than being continuously taught would still suggest that the longer, 2 hour long lectures resulted in poorer knowledge retention.
To do this, you need to search the literature for the impact of such treatment characteristics in a learning environment (e.g., How is memory affected by time? How is concentration affected by time? How does the type of material being learnt affect memory and concentration?). In answering these and other questions about time and learning, the literature will help you to understand how to set your research strategy (i.e., STAGE SIX); for example, the different lecture lengths you should study in your dissertation, the different subject material that participants should be taught, and so on. These are things that are discussed at more length in STAGE SIX. The important point at this stage is that you need to search the literature for books and journal articles that help you to understand these different treatment characteristics, and the potential impact that they could have in your dissertation.
When taking on Route B: Generalisation, you should also search the literature to see whether your main journal article has already been generalised in some way, as well as recognize whether your main journal article is, in fact, a generalisation of another study. This is important irrespective of the approach that you take within Route B: Generalisation. After all, some research, especially research that has produced a reliable measurement procedure (e.g., a reliable 22-item survey on service quality), will already have been replicated in another population, context/setting, treatment, or time. You should be able to find such research either by looking closely at the introduction, literature review, research strategy or reference list of your main journal article, or looking for those sources that have cited your main journal article (i.e., as we discussed in STEP C).