 In this video we're going to talk about qualitative data analysis software, sometimes called QDAS software or CACDAS computer assisted qualitative data analysis software. We're going to talk about some of the things that it can do, some of the things it can't do, some common misconceptions about using qualitative software. We're going to look at some of the different packages which are available for doing qualitative analysis with software and some of the differences between them. So the first thing we're going to talk about is the fact that computer assisted qualitative data analysis software is not analysis software. It doesn't actually analyze qualitative data for you. It's just a tool to help you do the analysis yourself. But it can be a really useful way to help you manage qualitative data, to manage your analysis process. I would describe it as a tool that helps you through the data collection, but especially the data analysis journey. And I'll talk a little bit about what I mean by that journey in a minute. So there are a lot of different qualitative software packages out there and there are important differences between them, but they basically all have the same features. They allow you to keep the data together. So you can import it from different sources, you can categorize those different sources of data, you can set what's called a tribute synom vivo or properties for those different sources that let you filter by different types of sources. And they also help you manage all those different sources of data and allow you to search across them. Some of them have advanced tools for allowing you to bring in data from different sources like social media. But most of them, all of them allow you to work with text data. Some of them will also allow you to work with multimedia data like audio, video or pictures. All the qualitative software has tools that let you annotate the data. So let's keep notes on it, comments and other ways of doing reflexive writing and analysis through writing. And the software can also help you manage those sources of reflexive writing and keep them together with your data. They also all allow you to create codes or themes. They do that in slightly different ways. None of them can suggest codes that should be in the data. That's still a process that you have to do yourself. And then they help you add data, so code data to those themes. So add selections of the data and put them towards different themes. And that's the main thing that qualitative data software is useful for. It helps you find and retrieve quotes, highlights, sections of the data that you've assigned to a different theme or a code. And that's something that you'll have to do quite a lot during most of the analysis processes that you undertake. And it can really help speed up the process if you use software to do that. Basically, because it's just helping you find all the different sections of data that you've already highlighted and assigned to those quotes. The qualitative software also allows you to search through the data. So you can see data by theme, by source type. You can see data that you've coded, data that you haven't coded. And you can also do keyword search if you're looking through text data. And a lot of software like Quercos, for example, have a built-in synonym database to help you do that. The software will also let you do visualizations of the data. So it might provide different tables, different graphs, different summaries of the data that show how the data compares between different sources, how you've coded it, what's under different codes. And then also maybe kind of visualizations as well. So for example, in Quercos and other software, there are word clouds to show how often different words appear in the sources. What does qualitative analysis software do? Well, basically, it's just a way of helping you analyze your data by keeping track of your data, your coding and the different analytical steps like notes and annotations. And this is why I talk about it helping as a tool through the analytic journey. Coding qualitative data or even analyzing qualitative data in any other method doesn't necessarily become a very strict linear process. There are lots of different ways you can approach it. It often becomes quite a sickly process where you're going back to the data, reading it and exploring it in new ways. And when used properly, qualitative software is a way that will help you explore that qualitative data, which can be quite overwhelming when you're presented with all of it. It's very in depth, very detailed, rich data, and it can help you find the key themes which will help you answer your research questions. A lot of people do qualitative analysis using pens and highlighters, printing out transcripts. But qualitative software has a bunch of advanced use over that. If you're doing it on paper, it can be a very tactile process, which a lot of people like. But it also means that when you've done your analysis, it's quite time consuming to go back and find the different quotes and highlights when you want to pull them out to write up or to support one of your research questions. Or when you start later on trying to turn your codes into themes and kind of like higher level order of understanding. It's much easier to search and find things in software compared to paper methods. It also has advantages that all software has such as undo and redo. So you can change your mind. And that's very difficult to do with paper and highlights. Once you've coded it, it's very difficult to go back. You can also code one piece of text to more than one code or theme. So if something is coming across different categories, it's easier to show that with them with kind of like hatched ways of merging different colors with different highlighters. In the software, it's also very easy to see a kind of timeline of what you've done. Go back and see snapshots, save snapshots of your data in different ways. And thus it's a lot easier to go through and do different stages of analysis compared to doing it on paper where you'd have to kind of print it out and start all over again. You can save project files. You can try different methods of analysis. It makes it a lot easier to go back and see data at any particular time during the process. You can also argue that the software also makes it a lot easier to keep backups of your work as you go along. Obviously, computers can crash that just as valuable as losing or having your hamster printouts of paper. But if you're doing proper backups, if you're emailing files to yourself, saving on the cloud or something like that, it should be a lot easier to make sure that you don't lose your data. Once you've been through and done the analysis, you've invested a lot of work in that and you don't want to lose that at any stage. The software also has other uses. So for example, you can use it for literature reviews. We've got video tutorials which show you how you can do that as well. But whenever I'm talking about qualitative software, I'm also very keen to talk about what it can't do in the limitations. A lot of people come to qualitative software thinking it's going to do the analysis for you. It can't. It doesn't understand your research questions. It doesn't understand what you want to do. And qualitative research, unlike statistical quantitative research, doesn't usually have easy one-click solutions that will show you summarize or pull out main findings from the data. It's something that you very much have to use your own analytical brain and your training and your experience with the data to understand. Qualitative software can help you do that, but there's no one big button to press that will analyze the data for you. Now some software has begun to introduce what they call auto coding features. But this is a very basic kind of descriptive level of coding. And it's based on machine learning and AI that is fallible and based on very simplistic interpretations of the data. It's rare that it's useful for any more than a very kind of quick look through the data. The same goes for keyword analysis. If you're just looking to see how many times somebody says the word angry, you'll miss the times people say livid or extremely agitated. So you have to be very careful. Really, the software is never a shortcut for going through and reading the data yourself. The software may help speed up the process of doing code and retrieval, but it's still a very time consuming process to do properly. You need to read through the data very closely. You need to understand it. And then when you go through and assign different sections of the data to different codes and themes, that's a very time consuming process. The transcripts of interviews or focus groups can be very long when you've got a dozen sources or so, it can take you days or weeks to go through and analyze. It can't make that process very much faster from a cognitive point of view, but it can make it faster to find those bits of data afterwards. The other thing to bear in mind about qualitative analysis software is that it can't write up your findings for you either. So just as a word processor can't write up your findings, it can't write your paper or thesis. The qualitative software can't write your conclusions for you. It can't analyze or interpret the data. That's still a job that you have to do, but just in the way that word processor helps you structure spelling mistakes and do other things that help you with the writing up process, qualitative software helps you through the analysis process. Now there are a lot of common misconceptions and criticisms about using qualitative software. Some of these are valid. Some of these have become a bit overblown. You can read a summary of these in a great article by Jackson Paulson-Wolf, 2018, and it's talking about zombie arguments, so arguments which keep being resurrected but really should be put to bed by now. We're going to talk about some of those here. The first one and the one you'll hear most often is that the software removes you from the data. It stops you from being too close to the data. Now that's a good argument, but I think it's something that is a criticism of how the software is used rather than the software itself. You can use it in a way where you start to rely too much on coding. You start to rely too much on just reading the data that you've coded. But there's a really great quote here from Barry that I want to read in its entirety because it really kind of debunked some of that. So Barry says that some of those who express this concern have often not used the software. Those who have tried the software have realized there's not possible to analyze your data without reading and being familiar with it first. In the worst case, it is possible that a researcher could only read the data in context during the process of initial coding. After that, they could just read snippets of the data that have been coded under each category to develop a final analysis without ever returning to the full contextualized data. However, this strategy is equally possible for those that use index cards, scissors and photocopies or word processor cut and paste functions. It may be possible to produce analysis using this superficial brush of the data, but it is unlikely to yield a quality analysis. And the number that is that, yes, it is possible to use software to not read the data, but really you should be using the software to help you read the data and read it more and understand it better. Some of the other criticisms around this are based on very old software where it was very complicated to use, especially in the 1980s when a lot of this software initially came out. It was very difficult to code things to data. There were a lot of steps involved. It required understanding a lot of Boolean logic and algebra. That's not the case anymore. So it's quite easy to use the tool just to focus on the data and not using the software itself. And that's really the essence behind Quercos as well, to try and make the software kind of get away into the background so you can focus on reading the data and reading the data better. Another argument is that the software can actually be quite addictive, especially creating more codes, coding to those codes, and you get kind of focused on that mechanical process of doing coding rather than understanding and interpreting the data. And again, that's definitely a risk and it's something that good users should be aware of. So just because the software allows you to create hundreds or thousands of codes doesn't mean that that's a good idea. Just because you can code everything to lots of different codes doesn't mean that's a good idea. If you're starting to get a lot more codes than you can really keep in your mind, then you should probably be reevaluating that. One of the good reasons that people turn to highlighters is that there's a limited number of highlighters and it actually kind of restricts, especially from an initial kind of coding and initial kind of read through the data, how many different types of codes and themes you're trying to pull out. But of course you can use software in the same way. If you restrict yourself to four or five codes or maybe a dozen kind of key codes or themes the first time you read through the data, you'll be using it in the same way and you can start to resist some of the temptations to code everything to a different theme. Another criticism which is very well connected to this is that the software forces you to code. But that's not true. The software allows you to work with the data and do any kind of analysis that you want to do on it. You don't have to do coding, you don't have to create codes and themes. You can do line by line coding, you can do IPA, you can do in vivo, you can do lots of other approaches which are just about reading the data and there's no reason that you have to do coding at all. You can even just use it as a tool to keep track of your notes and your comments and your own reflexive writing. Another criticism based again on the coding is that it results in quantification of the data. What that means is reducing things to just a simple number. So how many themes you have or the number of codes that you have in a theme. The most good qualitative analysis wouldn't rely on those kind of things because when you're doing qualitative analysis just because somebody has said something that doesn't mean that it has the same weight as when someone else says something. You really need to look into the depth of the code and what people are saying, the actual detail of the text itself. So though the software does allow you to count the number of times you've coded something to a particular theme, that doesn't mean that one theme is more important than another. Queer costs in particular tries to stop you doing that. By default, it doesn't even show you the numbers that you have coded to a different thing. It gives you a relative size so you can see what's emerging, but to put on the actual numbers that you've coded to a theme that's an optional extra. Another criticism is that qualitative software forces you to do a particular type of analysis. A lot of people would say one particular package is designed for grounded theory. Usually that's not the case. Usually you can be very flexible with these tools. There are different ways to use them. It may be that there are tutorials and examples out there which use a particular approach like grounded theory or code work analysis. But usually all the tools are flexible enough that you can have a different approach with them. Again, it's something which is easily done if you want to be led by the software. But if you take the proper approach and decide ahead of time as the five level QDA approach does, the basic tenets of this is that it encourages you to go back and decide what kind of analysis you want to do before you choose a software package or before you start using the software. That way you're not kind of blinded by any flashy features and you actually use the software in the way that you want to rather being led by a particular tutorial or a particular feature of the software that looks exciting. So there are a lot of different qualitative software packages out there and a lot of people have their own favorites. So what are the differences between them and why would you choose one over the other? Well, basically they all have the same basic features, but they all have different layouts. They will have certain different capabilities. So some might be better at working with certain types of data than others. For example, bringing in social media data or particular kinds of multimedia data. So the best thing is really to try, look for some reviews and see what's going to work best for you. I generally put the qualitative software into three categories. The first ones are the big three. So these are MaxQDA, AtlasTI and Envivo. These have all been around a while. They have the largest market share by far and they all have pretty much the same features. So they all do multimedia allowances so they let you bring in audio data and pictures. They have a lot of mixed methods and quantitative capabilities and features in there as well. So what it basically comes down to is the difference in layout, the difference in how they kind of conceptualize and describe the data and also what access you have. So some of them work on Mac and some of them don't. For example, you may have a license already at your institution for some of them and not. If there is a particular niche feature than you might choose one over the other. But most people would choose them for the layout. They will look a little bit different and work slightly differently with the data and it just becomes a kind of personal preference rather than any kind of particular methodological preference. The next categories for me are the Indies, like the Indie software. So these would be deduce, Transana, Quercos, F4 analyzer as some of the main ones there. Now, these all have kind of particular niche features that set themselves apart. So for example, deduce is cloud based, which is very useful if you're working collaboratively with people. Transana has extremely excellent video and multiple video stream functionality. Quercos is designed with a very visual approach. It's much simpler and easier to use but has fewer features. And F4 analyzer is designed around writing, so annotating and doing analytic text. You may also choose ones which have cloud based storage if you need to work collaboratively. So I mentioned already that deduce has that functionality. Quercos does now as well. So if you need to collaborate with other people, then those are probably the easiest ones to do because they allow you to work on the same project at the same time, no matter where anyone is. The other consideration is which software platforms they'll run on. So most of them will support Windows, some will support Mac. Very few support Linux, F4 analyzer and Quercos do. A lot of people also ask about open source alternatives. Often when people say open source what they mean is they want something that's free. And open source doesn't necessarily mean free, it just means that the code behind the software is freely available. So for example Transana is open source but there's a fee to use it. There are other alternatives like RQDA, which is a plugin for the R statistical software, ACQUAD and TAGET, which is quite a new one, which is cloud based. So there are pros and cons to these, that they may be cheaper or even free to use in some cases. What you would see is that there may be also less support available. The support for these is based on volunteers for the open source community. There may also be less documentation, less tutorials and less guides. And some can be quite difficult to install. So RQDA for example, there are a lot of dependencies and it can be quite difficult to get running on Windows at the moment. So all the software packages that I've mentioned today support the REFI QDA standard. And what that means is that you can save your project in that standard and then open it and save it again in any other qualitative software. So basically you can bring your coded data to and from any of the qualitative software packages that I've mentioned today. So Quercos and all the others now support that. And that's really useful because it means that you can work with other people who want to use a different software package. It also means that if there's a particular tool you want to use in one software, then you can bring your data into that software and use it just for that tool, even if you don't like the other ways that it works. And it also stops people from being locked down into a particular format, which was definitely a problem with some of the other software packages. Now, if you're looking for independent reviews of qualitative data analysis software, I would point you in the direction of the Surrey CACDAS network. It's run by the University of Surrey, but it's commercially independent and it offers reviews and training and guidance on all different software packages we've mentioned today. So my closing point is that qualitative analysis software can be an extremely useful tool when used properly. Most of it is quick and easy to learn. So all of them have at least a free trial. So I would recommend that you try a couple that look promising to you and see which one works out best. Speaking of that, you can download a free trial of Quercos from the link below. You can try for free a cloud version and the offline version, which are identical in the feature set. And you can see how Quercos makes that process very simple and intuitive as long as you only need to work with text data. But anyway, I hope you found that this tutorial was useful to you and gave you a bit of an overview of what you can do with some of the software packages. Do follow us and subscribe. And there are a lot more tutorials available on our YouTube channel.