ROUTE #1: Getting Started
ROUTE #1: Chapter-by-Chapter
Route C: Extension

In all approaches to Route C: Extension, you need to think carefully about all aspects of research quality - internal validity, external validity, reliability and construct validity - because the changes that you make to the research strategy of the main journal article - whether some aspect of the research design, research methods and measures, or sampling strategy - can significantly affect the quality of your findings. This reflects the extra level of originality and independent thought that goes into all approaches to Route C: Extension (i.e., compared with Route A: Duplication and Route B: Generalisation), especially design-based extensions and method or measurement-based extensions, but also population and context/setting-based extensions. These different types of extension are discussed in turn:

Population and context/setting-driven extensions

Population or context/setting-driven extensions require changes to the measures used within the research methods (e.g., the questions in the survey or the observation points in the structured observation) that were applied in the main journal article when studying the new population or context/setting in your dissertation. This is necessary because population or context/setting-driven extensions require you to add, modify or omit certain constructs and/or variables from the original study to reflect the differences in the characteristics of the new population or context/setting that you are studying. Such changes have a number of implications for the quality of your findings:

As you might have noticed, reducing selection biases when creating your sample (i.e., sampling biases) is particularly important to improving the research quality of dissertations that follow a population or context/setting-driven extension within Route C: Extension. As discussed in STEP FOUR: Sampling strategy, this is because the characteristics of the new population or setting/context that are important are likely to be different from those characteristics that were important in the main journal article. As a result, you have to rely less on the sampling strategy used in the main journal article and focus more on the population and setting/context that you are interested in. Therefore, when creating your sample in a dissertation following a population or context/setting-driven extension within Route C: Extension, you have to think: (a) what the most important characteristics of your sample are; and (b) make sure that the sample you select is as representative of the population you are interested in as possible (i.e., if you have to use a non-probability sampling technique rather than a probability sampling technique when creating your sample, you already know that the internal validity and external validity of your findings are being reduced because the representativeness of your sample is being threatened). However, since you have made changes to the measures used within the research methods, something that you don't do in Route B: Generalisation, you also need to make sure that the changes you have made do not make your measurement procedure do not reduce its reliability below acceptable levels (NOTE: We explain what are considered acceptable levels in the Data Analysis part of Lærd Dissertation).

Design, method and measurement-driven extensions

Design-based extensions require changes to the research designs used in the main journal article, whilst method-driven extensions involves changes to the research methods and measurement-driven extensions to the measures used within the research methods (e.g., the questions in the survey or the observation points in the structured observation). Of course, your dissertation may involve a combination of design, method or measurement-based extensions. As such, we deal with the broader implications of assessing research quality across these types of extension-based dissertations:

One of the overarching themes of design, method and measurement-driven extensions is that they require a lot more thought when it comes to assessing and ensuring the quality of the findings that are generated. Not only are there additional considerations when it comes to reducing threats to internal and external validity, but you have to focus a lot harder on ensuring the construct validity and reliability of your measurement procedure. At the undergraduate and master's level, you will not get this completely right; academics don't either. The main goal is for you to set a research strategy where the research quality of your findings is taken into account, and maintained as best as possible.

STEP FOUR
Determine how you will overcome such weaknesses in your dissertation

At the end of the day, the better you understand the weaknesses in your research strategy, the easier it will be to either overcome these, or manage them as best as possible. Remember that all research has limitations, which negatively impact upon the quality of the findings you arrive at from your data analysis. The best way to recognise these and overcome them is to get a good understanding of the five ways through which research quality is assessed; that is, based on the internal validity, external validity, construct validity, reliability and objectivity of the research [see the Research Quality section of the Fundamentals part of Lærd Dissertation to learn about these terms]. Reflecting on these will help you to reduce threats to internal and external validity, and improve the reliability and construct validity of your measurement procedure.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15