How Current CAT DILR Sets Compare to Pre 2015 Sets

DILR

Pick up an old CAT paper and the DILR section looks like a different exam. The data interpretation is recognisably data interpretation and the logical reasoning is recognisably logical reasoning, and they sit apart from each other in a way current sets rarely do.

That observation is real and it has consequences for how you use old material, which is the practical question underneath. Older papers are freely available and it is reasonable to want to use them, but they train a slightly different skill from the one the current section rewards. This piece sets out what changed, what did not, and exactly how to use pre-2015 material so it helps rather than misleads.

What Changed, Stated Carefully

Two things changed and they are worth separating from the general impression that everything is harder now.

The first is the blending. Older sets were more often cleanly one thing: a table with computational questions, or an arrangement with deductive ones. Current sets frequently require both in the same structure, so you build an arrangement and then compute across it, or read a data presentation and then reason about what it permits.

The second is the weight on selection. With a hard 40 minute sectional limit and sets that can absorb a third of it, choosing which sets to enter carries more of the outcome than it once did. That is a change in what the section rewards rather than in the content it tests.

What Did Not Change

More transferred than people assume, which is the case for using this material at all.

Since those four are most of what makes a candidate fast inside a set, old material is genuinely good for building them. What it cannot build is the part that changed.

What Old Papers Cannot Train

Two things, and both matter.

They cannot train selection under current conditions, because the section they came from had a different shape and different time pressure. Practising selection against a set of old sets teaches you to choose among options that are not distributed the way current ones are.

They also cannot train the blended structure, because the blending is exactly what was not there. A candidate who practises only on cleanly separated sets is fluent in two things that the current section frequently presents together, and the joining is its own skill.

How to Use Pre-2015 Material Properly

Purpose	Use old papers	Use current material
Building a family's recognition	Yes, ideal	Also fine
Representation practice	Yes	Yes
Data reading accuracy	Yes	Yes
Blended set handling	No	Yes, only here
Selection under timing	No	Yes, only here
Full sectional practice	No	Yes

The division is clean. Use old sets for type-wise volume, where you work one family at a time and the goal is fluency. Use current material for anything sectional, where the goal is choosing and pacing.

The Order That Makes Both Useful

Type-wise work first, mixed and sectional work second, and this is where old papers fit naturally.

Working through one family at a time builds recognition, which is what makes a cold start cheap. Old papers are well suited to this because their sets tend to be cleanly one type, so you can concentrate without the interference of blending while the family is still unfamiliar.

Then move to current sectional practice, where the families are not announced, the structures blend, and you have to choose. That second phase is where selection and pacing get built, and it cannot be done with old material.

Building the Blend Deliberately

If the blending is the genuinely new skill, it is worth a specific drill rather than hoping it arrives.

The habit that covers most of it is reading every question in a set before building anything. Blended sets punish narrow structures: you build a grid that answers the first question, then find the third question needs a computation your grid cannot support, and you rebuild. Thirty seconds of reading the questions first tells you what the structure has to hold.

The second habit is deliberately practising caselets and data-heavy sets, since those are where the blend most often lives. Candidates who trained on old material tend to be strong on pure arrangements and weaker exactly where the current section has moved.

Is It Actually Harder Now

Worth addressing directly, because it sits underneath the whole comparison.

The honest answer is that the section is more variable rather than uniformly harder. DILR swings more between years than Quant does, in set types and in how demanding a given paper is, so some recent papers have been brutal and others have not. A candidate who sat one hard year and reads old papers afterwards will conclude difficulty has risen, and a candidate who sat an easier year will not.

What is defensible is that the section now rewards a wider skill set: the same deduction as before, plus selection, plus handling structures that combine reasoning and computation. More skills tested is not the same claim as harder content, and it points at different preparation.

What the Variability Implies

Since the section swings between years, planning around any particular expectation is unwise.

The frame of covering a fixed syllabus does not apply here the way it does to Quant. What transfers across papers is recognition of recurring families and the judgement to assess a set quickly, and those hold whether the paper is generous or brutal.

It also means a bad DILR section in a mock carries less information than a bad Quant section. Some of it is the paper. Judge your DILR on a trend across several attempts rather than on any one of them.

The Same Logic Applies to Every Old Source

Past papers are not the only dated material in circulation, and the distinction drawn here settles the wider question too.

Books written some years ago, chapter lists inherited from a senior, and set collections assembled before the blending took hold all sit in the same category. Their content is largely sound, because deduction has not changed, and their framing of what the section rewards may be out of date. A book that presents data interpretation and logical reasoning as two separate disciplines to be mastered in turn is describing a section that no longer arrives that way.

The test to apply is the same one: does this material build fluency inside a family, or does it claim to tell you how the section behaves? Take the first at face value and check the second against current papers. A chapter of arrangement puzzles from an older book is as good as it ever was. A chapter explaining how to allocate your forty minutes across a section that no longer looks like that is worth reading sceptically.

This matters practically because most candidates assemble their material from whatever is available rather than by design, and dated framing propagates quietly. Nobody notices that they absorbed an approach to the section from a source describing a different one.

The Practical Plan

Putting it together, old papers have a defined job and a defined limit.

Use them heavily in the type-wise phase, one family at a time, for volume and fluency. Their clean separation is an advantage there rather than a defect, because you are trying to build one thing at a time.

Stop using them once you move to sectional work. From that point the material has to resemble the current section, because you are practising selection, pacing and blended structures, and none of those can be trained on sets from a differently shaped exam.

And do not read their difficulty as a benchmark in either direction. An old set that feels easy does not mean you are ready, and one that feels hard does not mean you are behind.

The Summary

Two things changed. The clean split between data interpretation and logical reasoning largely dissolved, so current sets frequently require both in one structure. And the weight on set selection rose, because a single set can absorb a third of a 40 minute section.

Most of the skill transferred. The recurring families are the same, deduction works the same way, representation choice matters as much, and data reading accuracy is unchanged. That is why old material is worth using and why dismissing it as too easy is a misdiagnosis.

So give it a defined job. Old papers for type-wise volume, one family at a time, where their clean separation helps. Current material for everything sectional, because selection, pacing and blended structures cannot be trained on a differently shaped exam. And treat the section as more variable rather than uniformly harder, which means judging your DILR on a trend rather than on any single paper.

If your DILR practice is mostly old material, you are building solving without building selection, which is the half the current section weighs most. A CAT preparation strategy review will show which half is missing, and a personalised CAT preparation plan sequences the type-wise work before the sectional work.