Quickstart guide

Select from the topics below to get a quick overview of where to find options and how to set up a first model. For detailed information on each page of the GUI, please see information on GUI components.

Quick overview of the GUI

The program utilizes six pages to set up and perform an HLM or GLIM analysis. These appear in order at the top of the page in the sequence they would be used to set up a new analysis: MLC, Data, Models (including Settings), Graphing and Run. The diagram below shows the general order of operations:

The Welcome or landing page that opens when first starting the program offers the user two options. One can either read in information on a previous analysis by clicking the Old Analysis option, or start with a new analysis using the New Analysis option.

The Models page is used to specify the model to be fitted. It consists of two parts, Models and Settings.

On the Models page, the model to be fitted is set up (for a new analysis) or displayed (for an old analysis) as a set of level-1, level-2, and level-3 equations. The Models page must be completed after completion of the Data page, but before the Settings page is accessed.

The Settings page is used to specify the type of outcome variable, link function (if applicable) and other options for the analysis to be performed on the model as set up on the Models page. If a moderation analysis is to be performed, this page also allows selection of  focal and moderation variables. The Settings page must be completed after completion of the Models page, but before the Graphing or Syntax page is accessed.

The Graphing page is only active if the Interactions field on the Settings page has been completed. Currently, the program will provide graphs for moderation analyses only. This page is used to request a simple slopes plot, a confidence interval plot, or both. Various plot parameters such as titles, line styles, colors, etc. may also be specified here.

The Run page is used to save the generated model syntax, start the analysis, and obtain plots (if applicable). After completion of the analysis, output may be displayed as HTML (the default option) or in text format in a separate window on the Run page. Moderation graphs will also be displayed on the Run page.

Finally, note the trash can positioned in the top right corner – to clear all input and start fresh, simply click on this icon.

Setting up a first model

Using the options above, we illustrate how to set up a simple level-2 HLM model with 2 predictors from scratch. If you would like to follow along, get the data here.

First, we need to select the data and specify which variables to use. This is done on the landing page of the program.

To start, click New Analysis to start setting up a new model.

The Data page opens

Our data are in the CSV file quickstart.csv on a local HDD, so click Select file to browse for this file:

The data in the file are displayed in the first table on the updated Data page. To specify a model, information on the structure of the data and the variables to be used in the analysis must be supplied in the second table.

There are four variables:

The levels of the hierarchical structure are captured in the variable LEVEL2ID, so we select this as ID 2 by clicking the check box for this variable in the ID 2 line. As we want to use all the other variables in the model, we click the check boxes for all three in the Variables line as showed below:

Click Update, and the program will automatically evaluate the levels of the variables and update the page to

By comparing the allocation with the data in the first table, we can see that X1 and Y are indeed level-1 variables, changing in value from record to record. X2, on the other hand, has the same value for all records with LEVEL2ID = 1224 and is correctly identified as a level-2 variable. It is worth noting that the user can override this allocation by changing the check boxes for the variables and clicking Update again. To reset allocation to that determined by the program, the Reallocate Levels button may be used. The Data page is now complete, and we click on Models to start specifying the model.

The Models page is now used to set up the model equations.

The first step is to identify the outcome variable by selecting Y from the drop-down list associated with the Set Response option in the Level-1 model field as shown below.

The Models page is updated, displaying an unconditional model with Y as outcome. Y is by default assumed to have both fixed and random intercept coefficients. The model also includes a coefficient for the residual variation at level-1.

Next, the predictor X1 is added to the model by clicking on the Level-1 Variables field. A variable can be entered in one of three ways (uncentered, group mean centered or grand mean centered). Click on the label X1 and drag it into the level-1 equation as uncentered predictor before releasing the mouse.

The Models page updates to include a slope equation for X1 at level-2.

Finally, add the level-2 predictor X2 to the intercept equation using the Level-2 Variables field.

Note that a variable may also be removed from the model by simply clicking on the “x” next to the variable name in the equation. The centering of a variable may also be changed by moving the mouse over it to access the little pop-up menu below, on which an alternative form of centering may be selected.

As our model is now complete, we can move on to specifying the type of outcome variable we have using the Settings page.

Specifying outcome distribution type and other options

The basic model now complete, all that remains prior to running the model is to specify the distribution type and other options (when applicable). Click on Settings to access the Settings page.

The program will automatically select a distribution for the outcome variable that it thinks most appropriate. Here, it selected Normal (HLM), which is exactly what we want. As there are no other options that need selecting, we can run the analysis. Click on Run at the top of the page to go to the Run page.

Run the analysis

The Run page is used to start the analysis, display output in txt format and display graphs (when available).

For the current model, we can (optionally) save the model information to a syntax file with MLCJSN file extension. By default, the file name will correspond to that of the data file (quickstart, in this case). Click Run Syntax to run the model.

A Progress box opens on the Run page, displaying iterative information. Once complete, all output files are available for viewing or downloading from the Run page.