In sophieberkhout/RRP-datasimulation:

User Manual

General

It is recommended to open the app on a desktop and not on a mobile device.

The app consists of different tabs. To go to the next tab, you can click on the 'Next' button on the bottom right of the sidebar. If you want to go back or navigate through other tabs, you can click on the tabs at the top of the page. This manual will explain per tab what is required for you to fill in and how to do it.

Home

In the Home tab there is only one thing to fill in: your student number. It is very important that you enter your student number, as we will be able to recreate your data if you run in any issues.

Descriptives

This tab is to provide the descriptives of the data. It is necessary to fill in this tab first, since the next tabs are dependent on the design that is filled in here.

Sample size per group

This slider indicates how many observations (i.e., participants) will be in each group. For example, if you put in 50, and you have two groups, than you will have a total sample size of 100.

Age

In this part you will fill in the characteristics of the age variable. It is required to fill in the mean age. However, the minimum and maximum age are not. You can leave these two blank if not applicable. For example, if your experiment only includes adults and you expect your sample to consist mostly of students, than your expected mean age would be 21 and the minimum age would be 18.

Gender

For gender you can change how many females you expect to be in your sample. On default, the slider will indicate half of your sample size. If you want only females, than you can slide the slider to the right. Or, if you have only males, than you can slide the slider to zero (the left). As you can see, the corresponding number of males is also shown.

Design

This is an important section to fill in. According to your assignment, you had chosen between two designs: two independent conditions/groups and three repeated measures or three independent conditions/groups and two repeated measures. Since the design influences what your data will look like, it is important to fill this in before continuing further.

Dependent Variable

This tab is about the dependent or outcome variable of your design. For example, if you test the effect of mindfulness on depression, than depression is your dependent variable. According to the assignment, you have already picked a measurement scale for this variable.

Variable name {#name}

This is the name of the dependent variable. Generally, you would use an abbreviation of the measurement scale. For example, if you measure depression with the Beck Depression Inventory you would fill in 'BDI' here. It is recommended to use a recognisable name, as this will be used as the column names in the dataset.

Measurement moment

For the measurement moment you fill in how to define these moments. Depending on the design, you will fill in either two or three names here. The default values are 'one','two', and, (if applicable) 'three'. It is not necessary to change this, but you can if you want it to be more recognisable for you. For example, if you have an experimental design with an intervention and you measure before and after, you can name these 'before' and 'after', or 'pre' and 'post'.

Group names

The group names are the names of the experimental conditions. These are the different groups that are in your design. So, for example, a control and a treatment condition.

Expectations {#expec}

Your expectations are probably the most important to fill in. This will determine what your data will look like. Note that the figure on the right shows how the data is currently. Changing the expectation values will also change the figure.

What you should fill in here are the expected mean values of the different measurement moments, per group. Take for example, the BDI. This is a scale ranging from 0 to 63 where values between 20-28 indicate a moderate depression. The higher the score, the more depressed a person is.

If you want to test whether depression goes down given a certain intervention, you will maybe want to test people that are moderately depressed. So, at the first measurement, you expect them to have an average score of let's say 25. Imagine you have a control group and an intervention group. You expect the score of the first measurement to be the same for both groups (randomised controlled trial). However, while you expect the control condition to remain relatively the same, as they do not get an intervention, you expect the intervention condition to go down.

Thus, the expected means for the control group will be around or equal to 25, and the expected means for the intervention group go from 25 to lower values.

Restrictions {#restr}

The restrictions indicate the minimum and maximum values of the measurement scale. If your scale does not have these values, it is not required to fill this in. If we move forward on the previous example, you see that the minimum of the BDI scale is 0 and the maximum is 63.

Manipulation Check

The manipulation check is to test whether your intervention actually worked. When you test the effect of mindfulness on depression, you probably have an intervention increasing mindfulness. Now, the manipulation check should measure if the mindfulness actually increased for the participants receiving the intervention.

Variable name

See previous section.

Relationship with 'dependent variable'

This part is about how the dependent and the manipulation variable are related to each other. If you want to test if mindfulness influences depression, these variables are expected to correlate. Thus, for example, if you expect depression to go down as mindfulness goes up, the correlation should be negative.

Expectations

See previous section.

Restrictions

See previous section.

Extra

According to the assignment you may add one or two extra variables to your design. You can choose a categorical and/or a continuous variable. However, it is not required to do so.

Categorical

Click on the categorical checkbox if you want to add a categorical variable. If you later decide to remove this variable, you can uncheck the box.

The categorical variable is a nominal variable consisting of multiple categories. For this variable you have to provide how many categories there are. For example, if you want to look at difference in hair colour, you could add three categories: blond, brown, or other. These are to be filled in for the names of the categories. Moreover, each category has its own probability. If you expect most to be either brown haired or blond haired, you could set the probabilities for both of these .4, and for the other category to .2. Note that these are probabilities, so logically, they should add up to one.

Continuous

Click on the categorical checkbox if you want to add a categorical variable. As with the categorical variable, you can uncheck the box if you later decide to remove this variable.

The continuous variable looks a lot like the age variable. Also for this variable, you provide a mean, a minimum, and a maximum value. What is different here, is that you can enter a relationship between this variable and the dependent variable.
For example, if you think hours of sleep per day is related to depression and your are interested in this relationship, you can add the variable sleep. Let's say that people with depression tend to sleep a lot. For the mean we could say eight hours of sleep, and the minimum is zero with a maximum of 24 (although very unlikely). Thus, more sleep indicates more depression, so we expect a positive relationship. If we expect it to be a strong relationship, we can set it to .7, for example.

Download Data

This tab is rather straightforward. In the main panel on the right you can see a table of the data to be downloaded. On the side panel you can decide if you want to download a .CSV file or an SPSS data file (.SAV). If you are going to analyse your data in SPSS, it is recommended to download the .SAV file. If you would like to use a different tool, like R, you can download the .CSV file.

sophieberkhout/RRP-datasimulation documentation built on June 24, 2019, 3:42 p.m.