Speaker 1: In this video, we're going to explain exactly how to write up the results chapter for a quantitative study, whether that's a dissertation, thesis, or any other kind of academic research project. We'll walk you through the process step by step so that you can craft your results section with confidence. So, grab a cup of coffee, grab a cup of tea, whatever works for you, and let's jump into it. Hey, welcome to Grad Coach TV, where we demystify and simplify the oftentimes intimidating world of academic research. My name's Emma, and today we're going to explore the results chapter, which is also sometimes called the findings chapter in a dissertation or thesis. If you're new here, be sure to hit that subscribe button for more videos covering all things research related. Also, if you're looking for hands-on help with your research, check out our one-on-one coaching services, where we help you craft your research project step by step. It's basically like having a friendly professor in your pocket whenever you need it. If that sounds interesting to you, you can learn more and book a free consultation at www.gradcoach.com. All right, with that out of the way, let's get into it. Before we get into the nuts and bolts of how to write up the results chapter, it's useful to take a step back and ask the question, what exactly is a results chapter, and what purpose does it serve? If you understand both the what and the why, you'll have a much clearer direction in terms of the how. So, what's the results chapter all about then? Well, as the name suggests, the results chapter showcases the results of your quantitative analysis. In other words, it presents all the statistical data you've generated in a systematic and intuitive fashion. The results chapter is one of the most important chapters of your dissertation because it shows the reader what you found in terms of the quantitative data you've collected and analyzed. It presents the data using a clear text-based narrative, which is supported by tables, graphs, and charts. In addition to presenting these findings, it also highlights any potential issues you've come across, such as statistical outliers or unusual findings. But how's that different from the discussion chapter, you ask? Well, in the results chapter, you only present your statistical findings. Contrasted to this, in the discussion chapter, you interpret your findings and link them to prior research, in other words, your literature review, as well as your research objectives and research questions. Therefore, the key difference is that in the results chapter, you present and describe the data, while in the discussion chapter, you interpret the data and explain what it means in terms of the bigger picture. Let's take a look at an example. In your results chapter, you may have a plot that shows how respondents to a survey responded to a survey, the number of respondents per category, for instance. You may also state whether this supports one of your hypotheses by using a p-value from a statistical test. In other words, you're just presenting the facts and figures. Contrasted to this, in the discussion chapter, you will say why these statistical findings are relevant to your research question and how they compare with the existing literature. In other words, in the discussion chapter, you'll interpret your findings in relation to your research objectives. Long story short, the results chapter's job is purely to present and describe the data. So keep this in mind and make sure that you don't present anything other than the hard facts and figures. This is not the place for subjective interpretation. Now, a quick caveat. It's worth mentioning that some universities prefer you to combine the results and discussion chapters. Even so, it's still a good idea to separate the results and discussion elements within the chapter, as this ensures your findings are both described and interpreted in a consistent fashion. Typically, though, the results and discussion chapters are split up in quantitative studies. If you're unsure, chat with your research supervisor to find out what their preference is. All right, now that we've got that out of the way, we can look at how to write up the results chapter. Let's do it. There are multiple steps involved in writing up the results chapter for a quantitative study. The exact number of steps will vary from project to project and will depend on the nature of the research aims, objectives, and research questions. For example, some studies will make use of both descriptive and inferential statistics, while others will only use the former. So, in this video, we'll outline a generic process and structure that you can follow, but keep in mind that you may need to trim it down based on your specific research objectives. The first step in crafting your results chapter is to revisit your research objectives and questions. These will be, or at least should be, the driving force behind both your results and discussion chapters. During your statistical analysis, you will have generated a mountain of data, so you need to use your research objectives and questions to sift through this data and decide what's relevant. Therefore, the first step is for you to review your research objectives and research questions very closely and then ask yourself which statistical analyses and tests would specifically help you address these. For each research objective and research question, list the specific piece or pieces of analysis that address it. Keep this list handy as you'll revisit it multiple times as you craft your results chapter. At this stage, it's also useful to think about the key points that you want to raise in your discussion chapter and note these down. Every point you raise in your discussion chapter will need to be backed up in the results chapter, so you need to make sure that you lay a firm foundation there. So, jot down the main points you want to make in your discussion chapter and then list the specific piece of analysis that addresses each point. Having considered both of these areas, you should now have a short list of potential analyses and data points that you know need to be included in your results chapter in some shape or form. Next, you should draw up a rough outline of how you plan to structure your chapter. This doesn't need to be highly detailed, but you need to think about how you'll order the various analyses in your chapter so that there's a smooth logical flow. We'll discuss the standard structure of a quantitative results chapter in more detail shortly, but it's worth mentioning now that it's essential to draw up a rough outline before you start writing or you'll end up with a wishy-washy mess of information. This advice applies to any chapter, by the way. As with all chapters in your dissertation or thesis, you need to start your quantitative results chapter by providing a brief overview of what you'll do in the chapter and why. For example, you'd explain that you will start by presenting the sample demographic data to understand the composition and representativeness of the sample before moving on to X to understand Y and Z. This introduction section shouldn't be lengthy. A paragraph or two is more than enough. The aim is simply to give the reader a heads up about what you'll cover, not to provide a summary of the findings. Also, it's a good idea to weave the research questions into this section so that there's a golden thread that runs through your document. The first set of data that you'll typically present in your results chapter is an overview of the sample demographics. In other words, you'll give the reader an overview regarding the demographics of your survey respondents. For example, what age groups exist and how are they distributed? How is gender distributed? How is ethnicity distributed? What areas do the participants live in? Why is this important, you ask? Well, the purpose of this section is to assess how representative the sample is of the broader population. This is important for the sake of generalizability of the results. If your sample is not representative of the population, you won't be able to generalize your findings. This isn't necessarily a bad thing, but it's a limitation you'll need to acknowledge. Of course, to make this representativeness assessment, you'll need to already have an understanding of the demographics of the actual population you're interested in. So, make sure that you design your survey to capture the correct demographic information that you will compare your sample to. But, what if I'm not interested in generalizability, you say? Well, even if you don't intend to extrapolate your findings to the broader population, understanding your sample will allow you to interpret your findings appropriately, considering who responded. In other words, the demographic data will help you contextualize your findings accurately. For example, if 80% of your sample was aged over 65, this may be a noteworthy contextual factor to consider when interpreting the data. Similarly, if a large portion of your sample was skewed towards one gender, this would be an important contextual factor to note. Long story short, regardless of your intention to produce generalizable results, it's essential to understand and clearly communicate the demographic data of your sample. So, be sure to put in the time and effort in this section so that you can contextualize your findings accurately. Before you undertake your core statistical analysis, you need to do some checks to ensure that your data are suitable for the analysis methods and techniques you plan to use. If you try to analyze data that doesn't meet the assumptions of a specific statistical method or technique, your results will be largely meaningless. Therefore, you need to do some checks on your data before you jump into the actual analysis. Most commonly, there are two areas you need to pay attention to. The first thing you need to check is the reliability of your composite measures. When you have multiple scale-based measures that combine to capture one construct, this is called a composite measure. For example, you may have four Likert scales that all aim to measure the same thing, but are phrased in different ways. In other words, within a survey, these four scales should all receive similar ratings, assuming they are indeed measuring the same thing. This is called internal consistency. Unfortunately, internal consistency is not guaranteed, especially if you developed the scales yourself. So, you need to assess the reliability of each composite measure using a test. Cronbach's alpha is a common test used to assess internal consistency. In other words, to show that the items you're combining are more or less saying the same thing. A high alpha score means that your composite measure is internally consistent. A low alpha score means you may need to scrap one or more of the individual measures. There are tests other than Cronbach's alpha that can be used, and there's some hot debate about which one is the best, but we won't get into that here. The key takeaway is that you need to undertake some sort of testing to assess internal consistency, and you need to present those test results in this section of your chapter. Once you're comfortable that your composite measures are internally consistent, the next thing you need to look at is the shape of the data for each of your variables. What do you mean the shape of the data? Well, for each variable, you need to assess whether the data are symmetrical. In other words, normally distributed in a nice bell curve or not. This is important as it will directly impact what type of analysis methods and techniques you can use. For many common inferential tests, such as t-test and ANOVAs, we'll discuss these a bit later, don't stress, your data needs to be normally distributed. In other words, symmetrical. If it's not, you'll need to adjust your strategy and use alternative statistical tests. To assess the shape of the data, you'll usually assess a variety of fairly basic descriptive statistics, such as the mean, median, and skewness, which is exactly what we'll look at next. Now that you've laid the foundation by examining the representativeness of your sample, the reliability of your composite measures, and the shape of your data, you can get started with the actual statistical analysis. Finally. The first step is to present the descriptive statistics for your variables. As I mentioned, the descriptive statistics will help you assess the shape of your data. As I mentioned, the descriptive statistics will help you assess the shape of your data. But depending on the nature of your research, the descriptive statistics could also play an important role in directly addressing your research objectives and research questions. So, what are descriptive statistics? When we talk about descriptives, this usually includes basic statistics, such as the mean. This is simply the mathematical average of a range of numbers. The median. This is the midpoint in a range of numbers when the numbers are arranged in order. Standard deviation. This metric indicates how dispersed a range of numbers is. In other words, how close all the numbers are to the mean, the average. Skewness. This indicates how symmetrical a range of numbers is. In other words, do they tend to cluster into a smooth bell curve shape in the middle of the graph? Or do they lean to the left or the right? And lastly, kurtosis. This metric indicates whether the data are heavily or lightly tailed relative to the normal distribution. In other words, how peaked or flat the distribution is. If these statistics sound like gibberish to you, be sure to check out our video covering the basics of quantitative data analysis. I'll include the link below. When you're presenting your descriptive stats, using a large table to present all the stats for multiple variables can be a very effective way to present your data economically. This saves you a lot of space and makes it easier to compare and contrast the statistics for each variable. You can also use color coding to help make the data more easily digestible. For categorical data, for example, data that shows the percentage of people who chose or fit into each category, you can either just plain describe the percentages or numbers of people who responded to something, or you could use graphs and charts such as bar graphs and pie charts to present your data. There's no one size fits all approach here. In some cases, it will make more sense to just present the numbers in a table or a paragraph. For example, if there are only two categories. While in other cases, graphs and charts will be useful. For example, if there are multiple categories. A pro tip, when using charts and graphs, make sure that you label them simply and clearly so that your reader can easily understand them. There is nothing more frustrating than a graph that's missing axis labels. Keep in mind that although you'll be presenting tables, charts and graphs, your text content needs to present a clear narrative that can stand on its own. In other words, don't rely purely on your figures and tables to convey your point. Highlight the crucial trends and values in the body of the text. Figures and tables should complement your writing, not carry it. All right, so that covers the basics of descriptive statistics. Depending on your research aims, objectives and research questions, you may end your analysis here. However, if your study also requires inferential statistics, then it's time to get started on those. All right, on to inferentials. Unlike descriptive statistics where the focus is purely on the sample, inferential statistics are used to make predictions about the population. So this part of the results chapter is where things can get really interesting. Inferential methods, broadly speaking, can be broken up into two groups. First up, there are those analyses that compare measurements between groups, such as t-tests, which measure differences between two groups, and ANOVAs, which measure differences between multiple groups. For example, you could use ANOVA to assess the difference in average weight loss between three groups that adopted three different diets. The second type of inferential methods are those that assess relationships between variables, such as correlation analysis and regression analysis. For example, you could use correlation analysis to assess the relationship between the number of hours studied and test marks earned within a sample of students. Within each of these inferentials, some tests can be used for normally distributed data, in other words, symmetrical data. And some tests are designed specifically for use on non-normally distributed data. So it's important to make sure that you use the right analysis tool for your data shape. Remember, you would have assessed data shape in your descriptive statistics section, so make sure that you align your inferential approach with those findings. There are a seemingly endless number of analysis methods and tests that you can use to crunch your data, so it's easy to run down a rabbit hole and end up with piles of test data. Therefore, you need to be selective about which methods you use. Ultimately, the most important thing is to make sure that you adopt the analysis methods that allow you to achieve your research objectives and answer your research questions. So let those two factors guide you. As with the descriptive statistics, in this section of your results chapter, you should try to make use of figures and tables as effectively as possible. For example, if you present a correlation table, use color coding to highlight the significance of the correlation values, or present a scatter plot to visually demonstrate what the trend is. The easier you make it for your reader to digest your findings, the more effectively you'll be able to make your arguments in the next chapter. Right. With both your descriptive and inferential statistics presented, you should have now laid the foundation for your discussion chapter. So at this stage, it's a good idea to quickly revisit that list that you drew up in step one and make sure you've covered all the necessary data to support your research objectives and research questions. If you have, you're ready to take the next step. If your study requires it, the next component of your results chapter will be hypothesis testing. Not every study will need hypotheses, so don't feel like you need to shoehorn these in if you haven't got any. As with so many things in your dissertation, the need for hypotheses depends on your research aims, objectives, and questions. So what exactly is a hypothesis? Generally speaking, a hypothesis is a statement that expresses an expected difference between groups or a relationship between variables. Importantly, it's a statement that can be supported or rejected by a statistical test. In other words, it needs to be very specific and measurable and cannot leave any room for interpretation or subjectivity. For example, a statement like, there is a relationship between study hours and test marks scored could be supported or rejected by a statistical test. For example, correlation analysis. Contrasted to this, a statement like, ice cream is the meaning of life, couldn't be supported by a statistical test, as much as I wish it could be. So if your dissertation or thesis included the development of hypotheses in the literature review chapter, this section is where you'd present them once again and test them using your statistical data. If you want to learn more about hypotheses, check out our detailed post over on the Grad Coach blog. I'll include a link below this video. One last thing. If your research involved developing a theoretical framework, you can, at this stage, present that framework once again, this time populating it with the hypothesis testing data. For example, if you were developing a theoretical model of the antecedents of trust and your hypotheses related to the antecedent variables, you could present the model again, this time incorporating the test results for each variable. Right, with your quantitative analyses presented, it's time to wrap up your results chapter and transition to the discussion chapter. To conclude your results chapter, the final step is to provide a brief summary of the key findings. Brief is the key word here. Much like the chapter introduction, this shouldn't be lengthy, a paragraph or two maximum. In this section, you only need to highlight the findings most relevant to your research objectives and research questions so that you lay the foundation for the next chapter. Don't provide a lengthy recap of each and every section of results. Just remind the reader of the key takeaways and wrap it up. If you work through your results chapter step-by-step as we've discussed in this video, you should land up with a comprehensive presentation of your key data. Keep in mind that what we've discussed here is a generic structure. There's no one-size-fits-all. The exact structure and contents of your results chapter will be influenced by your specific research objectives and research questions. As I mentioned in step one, start by crafting a list that covers your research objectives and questions and mapping that to the various statistical analyses and tests you've undertaken. Revisit that list as you work through each section of your results chapter and you can rest assured that you'll be headed in a good direction. All right, so that wraps it up for today. If you enjoyed the video, hit the like button and please leave a comment if you have any questions. Also, be sure to subscribe for more research-related content. Remember, if you need a helping hand with your research, be sure to check out our private coaching service where we work with you on a one-on-one basis, chapter-by-chapter, to help you craft a winning dissertation, thesis, or research project. If that sounds interesting to you, book a free consultation with a friendly coach at www.gradcoach.com. As always, I'll include a link below. That's all for this episode of Grad Coach TV. Until next time, good luck. Grad Coach