## Using Generalized Linear Mixed Model to extract a predictive formula/graph

Dear statisticians/SPSS users,

A little background information first: As you may or may not know, Rudolph Flesch created a Reading Ease formula in 1948, to be used for easily scoring texts on their comprehensibility by using average sentence length in words and average word length in syllables as variables. I'm currently in the process of validating this formula, as the last validation attempt I found in the literature was done in 1952, and the few validation studies there have been, did not exactly reach agreement (with correlations ranging from r=.55 to r=.87).
To carry out this study, I got 25 participants to take five different, prevalidated tests (varying in Reading Ease score of course), and let them answer a number of multiple choice questions. From this dataset, I want to deduce a predictive formula (preferably including a graph) that can be used to predict the chance of a random person answering a question correctly if the REscore is given. To achieve this, I have to use the Generalized Linear Mixed Model, but I am not entirely sure I'm using it correctly.

I will show what I'm currently doing below the example of how my dataset is structured:

Right, I'm using the GUI since I'm terrible at using the syntax only.
In the Data Structure tab, I've set the ParticipantID and QuestionID as subjects, and the REscore as repeated measure.
In the Fields & Effects tab, I've set Response as target, and I'm using binary logistic regression (since I'm treating each question as a trial, either correct or incorrect).
QuestionID and REscore are listed under Fixed Effects, whereas ParticipantID has been selected in Random Effects (with only ParticipantID as Subject combination). In both cases, Intercept has been included.

Based on this, I have two questions: First, am I doing it correctly and if not, what should I change? Second, if I have done this correctly, should it take such an enormous amount of time for SPSS to process it? Using the info listed above, I'm currently running the model, but it's been running for a good 15 minutes now without results.

PS: I'm using SPSS v20.