Appendix A: Using Webplotdigitizer

Appendix A: Using Webplotdigitizer

Appendix A: Using WebPlotDigitizer Sometimes you have a figure and it is not necessarily easy to extract numbers from it. For this you need an application such as the WebPlotDigitizer. This can be accessed here: https://apps.automeris.io/wpd/. This programme is distributed under the GNU Affero General Public License Version 3. When you open it, you get a screen that looks like this: You then need to save the graphic and import it through the Load Image button on the right-hand side. You then navigate to the figure that you want to upload; you should have previously saved it as a graphical file. This will then appear in the win- dow in the application. Here you can see an example from a paper looking at Pneumococcal carriage among children under five in Accra, Ghana; five years after the introduction of pneumococcal conjugate vaccine, this particular figure shows antibiotic resistance against different serotypes of pneumococcus (Dayie et al. 2019). We are going to look at the bottom figure in detail here. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer 153 Nature Switzerland AG 2020 E. Purssell, N. McCrae, How to Perform a Systematic Literature Review, https://doi.org/10.1007/978-3-030-49672-2 154 Appendix A: Using WebPlotDigitizer After which you choose the plot type—in this case a 2D Bar Plot. You then need to click on Align Axis so that it knows the scale used for the bars. In this case, the scale is on the y-axis, so you click the top and bottom of the axis and give the application the range, which is figure (a) above is 0–100 and in (b) 0–20. Appendix A: Using WebPlotDigitizer 155 If you look at the top left-hand size, you can see an expanded view of Point 2 to help you do this accurately. Point 1 is at 0 and Point 2 is at 20. You then click on each bar to see how far up the y-axis each extends. Here you can see that we had a look at the first three bars, and then pressed View Data to see the actual number associated with each bar. There are multiple options to export the data as a .csv file or via the clipboard. You can also change the number of figures after the decimal point. References Dayie NTKD, Tettey EY, Newman MJ et al (2019) Pneumococcal carriage among children under five in Accra, Ghana, five years after the introduction of pneumo- coccal conjugate vaccine. BMC Pediatr 19:316. https://doi.org/10.1186/ s12887-019-1690-5 Rohatgi A (2020) WebPlotDigitizer - Extract data from plots, images, and maps. https://automeris.io/WebPlotDigitizer/. Accessed 16 Jul 2020 Appendix B: How to Do a Meta-analysis Using R How to Do a Meta-analysis in R This chapter is going to show you how to undertake a meta-analysis using a package called meta (with a little ‘m’) in a programme called R (big R). The good news is that all of the materials that you need for this are freely available; the not so good news is that R takes a bit of getting used to. However, when you have got used to it you will find it among your best academic friends as it has so many useful packages that do so many useful things, not just meta-analysis but advanced graphics, general statistics, data mining, qualitative analysis, spatial analysis, survey analysis….the list is virtually endless. Installing R To install R, go to the R homepage https://cran.r-project.org/ where you will find instructions for installing R on Linux, Mac and Windows. If you are using Linux, you will find the interface is dreadful, as it uses the Terminal, and even if you are using the other operating systems you may still prefer a nicer interface than you get with R on its own. There are a few of these, but the most commonly used and the one that you will normally see on YouTube videos is called R Studio https://rstudio.com/ (look for the free version). Other options are: rkward https://rkward.kde.org/, Tinn-R https://sourceforge.net/projects/tinn-r/ or Rattle https://sourceforge.net/projects/tinn-r/ Depending upon which you go for: plain R or one of the other options when you open the programme up, you will get a fairly blank screen and a cursor. Although you may find some buttons and menus, R is really at its best when you type code in the form of a script rather than using buttons and menus. There are two reasons for this; firstly, it means that your work is 100% replicable, and secondly it means that if you have a problem you can send your code to your supervisor/second author/ significant other and publish it with your paper making your work super-transparent. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer 157 Nature Switzerland AG 2020 E. Purssell, N. McCrae, How to Perform a Systematic Literature Review, https://doi.org/10.1007/978-3-030-49672-2 158 Appendix B: How to Do a Meta-analysis Using R Packages R does have some built-in capabilities; it can do t-tests, ANOVA and all those sorts of things from scratch in what we call ‘base-R’. If you have downloaded R, that is what you currently have. In order to do most of the many, many things that R can do, you need to download different packages. There are, at the time of writing, 15,359 different packages of which a number has the capability to do meta-analysis. You can see the full range of meta-analysis packages here https://cran.r-project.org/ web/views/MetaAnalysis.html. We are going to use one of these, a package called meta which has its own homepage here https://CRAN.R-project.org/package=meta. One of the links on that page leads to the meta Reference manual which you should have a look at, but if it looks like gobbledygook don’t worry, this chapter is here to guide you through it. Start R Start R by clicking on the appropriate icon; this will be the blue R if you are using Base R or the appropriate icon if you are using a different interface. The first thing to do is to download the packages that you want. You do this by clicking the Packages menu, then the Install packages option. Then scroll down until you find meta and click on it. You will be asked where you would like to download it from, choose somewhere near (or nice) and you should then see all sorts of things happening, don’t worry about it let it all happen. What You Will See If you are using Base R, you will start by seeing one single window, this is called the Console and it is where the results of your analysis will appear. Resist the temptation to type code here as it will not be saved. To type your code, you need the Text editor by going to File then New script and you then see a new box appear. This is the Text editor where you will type your code and from which your code can be saved for future use. It is most important that this is where you type your code from now on. If you are using R Studio, you will see four boxes when you first open the programme; your Text editor is on the top left and your Console on the bottom left. Things to Know About R 1. It is free, free as in completely free. For this, we should thank the countless people who wrote and maintain R and all of the packages that are available. 2. It runs on all major platforms, but the details vary. At work, I use Windows which has control and R as the method of running functions; it is different for Macs and different again for Linux which I use at home. One of the reasons for Appendix B: How to Do a Meta-analysis Using R 159 using programmes such as R Studio and the other options listed at the start is that they are the same across all platforms, you just highlight the code you want and press run. 3. R does what you tell it to and no more. If you tell it to do a meta-analysis, it will; but it will not print the results unless you ask it to. You can get the results by just typing in the name of your meta-analysis. 4. In the spirit of doing what you ask and no more, even though you have down- loaded a package it will not actually run unless you tell R to run it. The function for this is library(name of package). I think of this as being like going to a library to take out my favourite book (or package). If you think about it that this makes sense, you might easily have 100 packages downloaded; it would be very wasteful of your computer’s resources to have them all running all of the time. 5. While on the subject of running packages, some packages rely on other pack- ages to run. So meta uses metafor for example. This is known as a dependency. You need not worry about it as generally R knows what is dependent on what and will look after it.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    33 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us