Few quirks and problems and questions..

5.40K viewsGeneral Discussion
0

Hi folks,

I’ve been using Quantrix on and off for a few years now but have only recently really tried to deploy it seriously. Rather than using it for financial analysis, though, I’m using it for analyzing student marks across University modules and years (which does require a fair bit of chopping around analysis). But I’ve run into a few problem and I wonder if anyone can help me with these?

– At the moment I have a matrix for each module, with categories Student (row), Point (column – that’s “assessment point”, ie, the lowest unit of things that the students get marks for) and Year (filter), with Points grouped. One awkward problem is that Students are identified by three parts, a number and a name, but there seems to be no way to divide a category item into three parts, so at the moment I have to put all three of them into one item which makes them awkward to type and match. I can’t add extra columns without incorrectly listing “last name” and “first name” as assessment points.

– I can’t have one matrix for all modules because the structure of Points may be very different between different modules. But, is it possible for me to create tables that link together Student items from different matrices? Can I say, for example, “the list of items in the Student category of this (summary) matrix should always be equal to the list of items in the Student category of this (module 1) matrix plus the list in this other (module 2) matrix?”

– Because I have Year as a category, I usually use it as a filter to view only one year at a time. Because it’s rare (but it can happen) that the same student takes the same module in two different years, this means that I see blank entries for all the students who didn’t take the module that year. I know I can turn these off by selecting “Hide Blank Lines”. But if I do that then when I try to enter a new student by pressing Enter with a Category Item selected, the new student does get created but they don’t show up and I can’t edit them – because their line, being brand new, is blank and therefore hidden! Is this a bug? Is there a way around it?

– In most cases it’s useful to store the Maximum Marks available for each Point. In my old spreadsheet I had a special row for this at the top of the spreadsheet. But in Quantrix, if I create that row, then I am claiming that “Maximum Marks” is a student (because that’s the row category) and they’ll get included in analysis. Is there any way to decorate a category item in this way?

– Is it possible to make “do nothing” an option in an IF? Each student gets a Result Code which is calculated by a fairly complicated process from their final result. What I’d like to do is to have something like this, as separate formulae:
1. Code = If (Summary.Mark > 40 #and# Coursework.Mark < 30, “TRC”, <do nothing>)
2. Code = If (Summary.Mark > 40 #and# Exam.Mark < 30, “TRE”, <do nothing>)
The problem is that there doesn’t seem to be a way to say “do nothing”. If I put an empty string as the “if false” option then, in the above case, the formula for adding a TRE will overwrite anything but a TRE with an empty string. So I have to have all the potential Ifs packed into a single huge formula a la Excel.

– Quantrix is great for manipulating summaries like averages and standard deviations but I also often need to calculate a distribution. In Excel I just grab a few cells from the corner of the spreadsheet to put the distribution bin values and results in, but in Quantrix I can’t do that because there are no “spare” cells outside of the Categories. How do I create a summary that has multiple values in that way?

0

hyphz,

Don’t doubt – you have very similar issues other Quantrix users including myself have nearly every day. Quantrix requires a different mindset when dealing with data and formulas than when we are working in Excel. That’s not surprising as the conventional spreadsheet metaphor has been around us for 34 years now and it is not easy to switch to the multidimensional thinking in Quantrix.

I try to give you some hints but remember that better solutions are possible. All I am describing here is available in the attached model.

The first step is to define each category in separate matrix:
– Years
– Students
– Marks
– Points
– Code List

All these matrices are in the folder “Structure” in the Model Browser. You can open them quickly by switching to the “Manage Structure” perspective.

The issue with the Students has been solved by using the ID as the item name and then by defining a descriptor “Name” in the Student category.

I don’t have a better solution regarding adding a new item when “Hide empty rows/columns” is active:
1. Unhide all empty rows/columns
2. Enter the new items including the data
3. Hide empty rows/columns again

If you don’t want to get particular items included in the analysis you can use the SKIP statement in the corresponding formulas.

Example:

A=SUM(B) SKIP Maximum Marks

The “TRC” and “TRE” issue is solved by creating a Code List matrix where the codes are linked to the range values. The analysis matrix (switch to the Analysis perspective to the matrix “Mark Analysis”) then gets the codes by using one single formula including a lookup function.

The matrix “Marks Statistics by Student” in the same perspective shows how you a simple example of a statistical summary. You can add much more here.

I am fully aware that the model provided here is very simple and is not giving you a satisfying answer to all of your questions. But maybe it can help to show you some hints how you can model a Student Marks analysis in Quantrix.

Let me know if something is not clear or if you have further questions.

All the best,
Dominik

You are viewing 1 out of 1 answers, click here to view all answers.