Introduction

statscloud is a new statistics package that was designed primarily for students. Unlike other statistics software, which presume a certain amount of expertise, statscloud does not overwhelm students with lots of options, tickboxes or statistical jargon. It has a modern, intuitive interface that makes running and interpreting analyses much more straightforward.

They can use the app instantly just by visiting the link https://statscloud.app or, if they choose to, they can install statscloud as a Progressive Web Application so that it looks and functions just like a normal app.

There are no limits to access the app: There are no timeout limits, no session timeouts, no idle timeouts, and no restrictions on how many people can access it at a time. The app can also function completely offline, so it's accessible to students without a reliable internet connection. All of this means you can be guaranteed the app will always be available for you and your students.

These features make statscloud an ideal choice for students and, because of this, more and more universities across the world are starting to use statscloud in their teaching. While the initial version of statscloud is complete, there are more features on the horizon, so now is the perfect time to start introducing it to your students.

Note: If you're not yet familiar with StatsCloud, take a look at the Introduction here

Getting started

At the start of an academic year, lecturers often give students instructions on how to download and install a statistics package. This can sometimes be quite time-consuming and students sometimes run into technical difficulties doing so.

Another consideration is that a lot of students come through to university now without desktops or laptops, simply because they don't need them; they can often do everything they need to on a tablet or a Chromebook. However, because statistics software is usually not compatable with these devices, it is typically impractical or impossible to install statistics software on them.

With StatsCloud, neither of these things are an issue. Students don't need to download or install any software and, to open the app, they just need to be directed to https://statscloud.app.

Sharing projects

As statscloud is a web application, there's no need to have project files stored offline; you can open a new project with a single link and customise exactly what gets loaded when students open it. For instance, a link to the statscloud app can take students to a blank project, a project preloaded with a data set, or a project with some data analysis already conducted. You can disseminate these links to students throughout your teaching materials and ensure that, when they open them, they see exactly what you want them to.

Project links

Creating project links is easy to do. When you're using the app and have got the project to appear to students as you would like it to (e.g. after having entered data, set up variables or run an analysis), you can then generate a link to the project using the Share button in the app's navigation menu.

These links are very versatile and you can customise exactly what you'd like students to see when they follow them. For instance, the link can take them to a project with some variables already set up for them, some active filters in place, or with several analyses completed already. You can even choose which tab the project will open up in and what the project is called.

Because shareable links can be quite long, you can shorten them with a URL shortening service (e.g. bit.ly and can even generate your own QR Code through the same Share menu. These links could be disseminated to students in posts on your Virtual Learning Environment, in online chats, or through emails. QR codes could also be placed on PowerPoint slides so that students can scan them and open them up on their touch devices instantly.

Embedding

You can also embed the app in HTML documents, such as online articles, blogs, or posts in your Virtual Learning Environment. An example of this is below:

The embedded app functions just like the full app. You can interact with all the content inside the embedded app as if you were using the app in it's own window. You can also customise the appearance of the embedded app so it fits with the style of your document. Just like the actual application, the interface adapts to the size of the frame it is displayed in (e.g., it will move to the 'mobile' view when the width of the frame is made smaller) so you can be sure the embedded app will always be displayed well.

To get the code for an embedded app, simply navigate to the "Embed" tab of the same Share pop-up. You can then copy this content and paste it into any rich-text editor that allows HTML.

<iframe src="/project/temp/?data=marvel-movies&title=New%20Project" title="StatsCloud" allowfullscreen="true" width="100%" height="800px" title="StatsCloud" data-external="1" style="border: 0px; border-radius: 5px; box-shadow: 2px 2px 8px rgba(0, 0, 0, 0.2); min-height: 600px" ></iframe>

Embedding the statistics app in your teaching materials is a great way of making your content more accessible to students!

Using project templates

With project templates, you can ensure students have a project set up correctly. An example of when this may be useful is when, early on in their statistics training, students get confused between the format for data entry in an independent-samples design and a repeated-measures design. To help them understand the difference between them, you could show them an independent-samples design template and a repeated-measures design template:Score&title=Two%20related%20groups). This will ensure they enter the data correctly and will know how to do this themselves in future.

In addition, you can also provide a template for an analysis in these project links. As well as having the spreadsheet set up with variables ready for data entry, these projects can also have an analysis set up and ready to run too (for instance, like <a href='https://statscloud.app/project/temp/?vars=X,Y&dataTypes=1,2:r&analyses=:1,2:pearsonsR&title=Pearson's%20R%20Template' target='_blank' rel='noreferrer'>this correlation template). Once the data has been entered, students simply navigate to the "Analyes" tab and click Run test. This can be useful when introducing an analysis to students for the first time and ensuring they run the exact test you would like them to.

Note: A range of analyses templates is already provided in the Templates link of the home dashboard.

Linking to public data files

You can link a data set file (.csv) to statscloud so it can load up with the data set populated. When a data set has been linked to StatsCloud, the app will prompt the user to download the file first and then import it into the app. Once the data has been imported, statscloud will load up the project with the linked data set populated.

For this method to work, the data set should be hosted publicly online (e.g. https://raw.githubusercontent.com/dangurn/datasets/master/fisher-iris.csv. The link to preload this data set in statscloud would be: statscloud.app/project/temp/?data=https://raw.githubusercontent.com/dangurn/datasets/master/fisher-iris.csv.

To get a link to a project populated with another data set, the most simple thing to do is edit the link above, replacing the text after ?data= with a new file URL.

You can also get the link via the 'Share' window (as shown above). To do this, just open a blank project, and click the Import from URL icon in the toolbar or follow File > Import from URL menu. Once you have entered your URL and statscloud has loaded up the data into a new project, you can get the link to this project by clicking the Share button in the navigation bar.

Loading your own data files

You can of course import an offline data (.csv) file into StatsCloud. To do so, simply drag it onto the statscloud spreadsheet in the Data tab (if using a desktop) or click/tap the Import button. Like other statistics packages, though, this requires you to have your data files stored offline and requires all your students to have access to this file offline too. An alternative method is to store your data file at a public URL and set up a statscloud project to read it when the app loads up (as explained above).

The main advantage of hosting your own data files is that you have full control of them. When anybody opens a project with a data set linked to it, statscloud will always load up the newest version of that data set, so any changes you have made to it will be reflected immediately, and you can ensure all students see exactly the same data. This is particularly useful when students are working collectively on a data set you have provided for them, and saves you having to disseminate original (and updated) versions of the files.

If you'd like to make use of this feature, you'll just need to make sure your data files are hosted somewhere online so statscloud can find them when the app loads. A popular place to store files and teaching materials is GitHub. If you don't already have an account, it's very easy (and free) to set one up.

Data entry

When entering data manually, statscloud has a range of tools to make this much easier for students. When they start entering data, the app will select the best data type and assign labels to categorical variables for them automatically. Editing variables is much more streamlined too, and there are no other options for students to worry about.

Editing variables

Within the Data tab, students can click / tap a column header to edit basic features of their variables (including the name and data type), and can even colour code columns too to make them stand out and easier to find.

In addition, because many students will want to use touch devices where they can, statscloud has a new, touch-friendly Form view to make data entry on these devices much eaiser. They can toggle this on or off just by touching the switch in the top corner.

Setting up repeated measures designs

The method of declaring repeated measures designs in other statistics packages is quite cumbersome and students can often find it confusing. Typically, variables are given headings such as "confidence_smart" and "confidence_casual". However, these names can be very verbose (particularly when working with multiple factors). Not only this, students can be confused by how columns are treated as separate variables (rather than levels of factors), and how the predictor variable is never declared anywhere. Linking these variables together later when running analyses or creating charts can also be quite tedious and can lead to errors.

To remedy this, statscloud has a revolutionary new way of declaring repeated-measures designs that is much more intuitive for students. Students declare their one outcome variable as normal and split this into conditions using the 'Related Conditions' tool in the Variables tab.

In the Data tab, the conditions are nested neatly under the outcome name. This prevents the need to have long names, and makes the data set much clearer to read.

Another advantage of this method is evident when creating charts and running analyses. Instead of setting up factors and levels and then assigning columns to conditions, students simply select the outcome variable. As statscloud already knows there are levels are attached to it, it includes them in the analysis automatically so there is no need to assign them again.

Data cleaning

Before teaching students how to run analyses, it's a good idea to teach them how to clean a data set first. Other statistics packages can make this a little confusing for students, as getting 'descriptive' statistics for a variable requires them to go through the same process as running an analysis. In StatsCloud, though, data summaries are kept within the Data tab so to not confuse students, and get them used to the idea of spotting errors with their data sets before they start running any analyses.

In the Data tab, each column has a cell at the bottom which provides a summary of that column and warns students if there are any issues with it (e.g. if some data points are missing or the data is skewed). By clicking on this, students are able to see exactly what issues there are with this column so they know what problems they are likely to encounter when running analyses and what they'll need to do to fix them.

Task: Teach students how to clean a messy data set

• Give students a link to a messay data set
• Give them a few minutes to spot as many issues with it as they can (e.g. missing values, incorrect data, potential outliers)

Visualisations

One of the best ways of teaching students about data distributions is through visualisations. While statistics software packages are limited in what charting options they offer, statscloud allows students to generate advanced and highly-informative charts, such as raincloud plots, very easily. These charts present an excellent opportunity to teach students about box plots, normal distributions and data skewness, all in one place.

Tip: Introduce students to raincloud plots

• Give students a link to a raincloud plot
• Ask students to identify what the chart reveals about the spread of data across all groups.
• Using the editing tools, ask students to customise the chart in a format that would be appropriate for a scientific (e.g. APA) journal.

Teaching statistics theory

As statscloud was designed primarily as a teaching tool, it has some unique features to help students understand the theory behind their analyses as they use the app.

Live formulas

One of the most powerful features of statscloud is that it provides worked formulas for every statistic it calculates when running an analysis. This makes it possible to teach statistics theory within the app itself.

When you click on any statistic in an output table, you can view the template of the formula used to calculate as well as the formula with the relevant numbers plugged in.

Tip: Use the live formulas to help students understand statistics theory

• Give students a link to an analysis
• Ask students to do a hand calculation of a statistic (e.g. a mean standard deviation, or df value), then ask them to check their working by clicking on the value in the output table
• Demonstrate to students the difference in calculations across two similar tests. For instance, by toggling between the student t test and the Welch t test and note how the denominators change.

If you choose to, you can also use these formulas in your teaching materials. Microsoft Office allows you to paste in the LaTeX formulas generated by StatsCloud. To do this, copy the formula from statscloud by clicking the "Copy" button in the formula pop-up window. When in Word, click Equation from the Insert menu in the ribbon and Insert New Equation. You'll need to change the input to LaTeX in the conversions options, then click convert to 'Curent - Professional'. For more detailed instructions on how to do this, please see a guide (here)[https://superuser.com/questions/340650/type-math-formulas-in-microsoft-word-the-latex-way.

Teaching students to code

To help give students additional skills on their degree, some universities have taken the brave step of teaching students how to run statistical analyses via a programming language (e.g. R). While this can work well for some students, many would likely benefit from the more user-friendly interface provided by statscloud before migrating fully to a programming language.

The "Code" tab in statscloud was designed to do exactly this and is a great way of introducing students to a programming language. You can open up the Code tab at any point during your project (when you're declaring data, when you're sorting or filtering your data, or when you're running analyses) and everything you have done in statscloud so far will be reflected in the Code tab. By flicking backwards and forwards to the code at various points during your project, students can see how the code changes.

Students are able to explore the code in more depth by clicking on any line to view annotations to see exactly how it works.

Making lectures more interactive

Often, the best way for students to learn how to run a statistical analysis is to follow what the lecturer is doing live, step-by-step. This isn't always possible or practical to do in a lecture environment as it depends on all students having laptops with them. Instead, students will have to take notes on what steps to take and try to replicate these instructions later.

With StatsCloud, though, statistics lectures can now be more interactive. Students can open the app in the lecture on any device they have with them (phones, tablets, Chromebooks or laptops) using a link or QR code on the lecture slides, and follow what lecturers are doing on screen on their own device.

Online teaching

When running statistics sessions online, it makes sense to use a statistics package that runs online. By doing so, you won't need to worry about students having downloaded or installed any software before the session (or whether they're using the latest version); you can just provide them with a link to a project and be sure that they can all open the latest version of the app instantly.

While you're running your teaching session online, you can paste links to statscloud projects in a chat window. The links you provide can be customised so they load up projects with the exact content you want them to. This could simply be a blank project, a project preloaded with data, or a project with some data analyses already conducted.

By making use of StatsCloud's ability to load in data from a URL, there's also no need to send data files to students either; you can upload your data to a public repository before the session and generate statscloud project links that reference these when the app loads up. If you're in control of the data file (e.g. you have it uploaded on your own GitHub account), you can make changes to this file whenever you want and know that statscloud will always load up the latest version of it when the app is opened.

Ensuring that all students have the app open together on their own devices allows for more interactive teaching sessions too. Students can follow your instructions live instead of taking notes during the session and trying to reproduce everything on their own later. The features make statscloud an ideal choice for online statistics teaching and makes life easier for both educators and students.

Accessibility

It is possible that some of your students will have accessibility needs (e.g. visual or motor impairments) so it is important to get them set up with an app that is possible and intuitive for them to use. The statscloud app has been designed to be accessible to all users and is therefore suitable to recommend to students. The whole app can be used with just a keyboard, and it has been programmed to work with screen-readers too. In addition, it has a 'dark mode' to reduce eye strain, and all animations are turned off automatically when animations are turned off in the device settings.

Note: For more information about accessibility, please see our accessibility statement

Improving Student Experience

Students appear to prefer using statscloud over other statistics package as it's far more modern and intuitive for them to use, and very easy to get started with. Students are often quite apprehensive about statistics, but statscloud makes statistics much easier for them to digest. One good reason for introducing it to students therefore is to help improve student experience. statscloud can help improve this in the following ways:

• The app doesn't require students to buy or borrow a new device to run a statistics package
• Ensuring your students don't run into any technical issues when downloading / installing a statistics package
• Ensuring the app is always available for all your students, and is not subject to session timeouts, idle timeouts,
• Introducing an app to students which they find aethestically pleasing and similar in appearance to other apps they have used previously (e.g. Google Docs, Sheets .etc).
• Making sessions more interactive and engaging. Students learn best when doing!