UCAT: Abstract Reasoning

[swagatronium]

hey guys, interested in hearing how u would approach this Q. personally, found that there was a lot going on so i jotted down the relationship of the colours of the circle and triangle w/ the # of arrow type. This is what I wrote:

A: C - <-> Odd = black
T - (other arrow type) Even = black

B: T - <->
C - (other arrow type)

do u reckon i shouldve done it in my head, or is using the whiteboard the way to go with this pattern?

 

[mmorris]

hey guys, interested in hearing how u would approach this Q. personally, found that there was a lot going on so i jotted down the relationship of the colours of the circle and triangle w/ the # of arrow type. This is what I wrote:

A: C - <-> Odd = black
T - (other arrow type) Even = black

B: T - <->
C - (other arrow type)

do u reckon i shouldve done it in my head, or is using the whiteboard the way to go with this pattern?

View attachment 5168

I would have wrote this down aswell. I find it quicker to figure out if theres a relationship when the numbers are right in front of me so i would write something like:

Set A:
TOP LEFT BOX: White Circle, Black Triangle = 3 (pairs of arrows pointing in), 1 <->
TOP RIGHT BOX: Black Circle, White Triangle = 2 (pairs of arrows pointing in), 2 <->
BOTTOM LEFT BOX: White Circle, White Triangle = 1(pairs of arrows pointing in), 2<->

In timed practice though I would write it out shorthand to save time by keeping the arrow shapes in my head:
WC BT= 3, 1
BC WT= 2, 2
WC WT= 1, 2

By looking at those three relationships i can see that WC=odd (pairs of arrows pointing in), BC=even (pairs of arrows pointing in), WT=even <->, BT=odd <-> . I would then check this rule with one last box in set A very quickly to confirm its legit.

Then for set B i already know without looking at any boxes that the rule must be the opposite or very similar so just by checking the bottom right box in set B I can tell thats the case.

 

[swagatronium]

I would have wrote this down aswell. I find it quicker to figure out if theres a relationship when the numbers are right in front of me so i would write something like:

Set A:
TOP LEFT BOX: White Circle, Black Triangle = 3 (pairs of arrows pointing in), 1 <->
TOP RIGHT BOX: Black Circle, White Triangle = 2 (pairs of arrows pointing in), 2 <->
BOTTOM LEFT BOX: White Circle, White Triangle = 1(pairs of arrows pointing in), 2<->

In timed practice though I would write it out shorthand to save time by keeping the arrow shapes in my head:
WC BT= 3, 1
BC WT= 2, 2
WC WT= 1, 2

By looking at those three relationships i can see that WC=odd (pairs of arrows pointing in), BC=even (pairs of arrows pointing in), WT=even <->, BT=odd <-> . I would then check this rule with one last box in set A very quickly to confirm its legit.

Then for set B i already know without looking at any boxes that the rule must be the opposite or very similar so just by checking the bottom right box in set B I can tell thats the case.

Appreciate ur thoughts on it mmoriss

 

[mmorris]

Hey do you guys flag the complete the statement/complete the series questions and do them only when you finish all the Set A/b/ questions?

Since with the Set A/B questions once you figure out the pattern you've got 4 questions finished compared to just 1 question finished when you figure out the pattern for complete the... questions

 

[swagatronium]

Hey do you guys flag the complete the statement/complete the series questions and do them only when you finish all the Set A/b/ questions?

Since with the Set A/B questions once you figure out the pattern you've got 4 questions finished compared to just 1 question finished when you figure out the pattern for complete the... questions

i agree with ur thinking that it may be better worth for ur time, tho I think u should still go for the complete the statement Q’s. what I find personally works for me is just going through all the patterns and if I can’t get one pattern in perhaps 10-15s, I’ll guess, flag (the 1st Q if it’s a Set A/B/neither or Set A/B) and move on. I’ll flag the ones also I think I’ve only gotten the primary pattern for (and not the secondary). after I’ve reached the end of the AR section, I’ll go over the flagged ones one or (rarely) two more times then run out of time.

different strats could work tho so feel free to experiment ye

 

[mmorris]

i agree with ur thinking that it may be better worth for ur time, tho I think u should still go for the complete the statement Q’s. what I find personally works for me is just going through all the patterns and if I can’t get one pattern in perhaps 10-15s, I’ll guess, flag (the 1st Q if it’s a Set A/B/neither or Set A/B) and move on. I’ll flag the ones also I think I’ve only gotten the primary pattern for (and not the secondary). after I’ve reached the end of the AR section, I’ll go over the flagged ones one or (rarely) two more times then run out of time.

different strats could work tho so feel free to experiment ye

yeah i do the same thing with the flagging

I think personally my best strat would b something like:
1. go through all the Set A/B style questions and flag the ones I couldn't figure out the pattern for
2. skip complete series/statement questions
3. go through the flagged Set A/B style questions
4. repeat step 3 as many times as needed until I figure out all the remaining patterns
5. with the remaining time finish the complete series/statement questions

 

[swagatronium]

yeah i do the same thing with the flagging

I think personally my best strat would b something like:
1. go through all the Set A/B style questions and flag the ones I couldn't figure out the pattern for
2. skip complete series/statement questions
3. go through the flagged Set A/B style questions
4. repeat step 3 as many times as needed until I figure out all the remaining patterns
5. with the remaining time finish the complete series/statement questions

yeah that seems fine dude, only thing is i imagine u would have an eye on the clock for how many repeats of step 3 (ur step 4) u could be able to do. i sometimes find with the prep company im using (cant name because of TOS), that sometimes you can get partial patterns where you can get 2-3/5 or sometimes even 4/5 without figuring the entire rule or not being able to verbalise it, and just by judging between boxes in sets A and B. Other times, i find it a bit harder to do that especially when there's not a pretty blatant difference such as colour or number.

i imagine then u might want to try and do step 5 beforehand just to secure marks where u can if u have done a few rounds of step 3 and still are grappling with those ones, then worst case guess those ones.

 

[mmorris]

yeah that seems fine dude, only thing is i imagine u would have an eye on the clock for how many repeats of step 3 (ur step 4) u could be able to do. i sometimes find with the prep company im using (cant name because of TOS), that sometimes you can get partial patterns where you can get 2-3/5 or sometimes even 4/5 without figuring the entire rule or not being able to verbalise it, and just by judging between boxes in sets A and B. Other times, i find it a bit harder to do that especially when there's not a pretty blatant difference such as colour or number.

i imagine then u might want to try and do step 5 beforehand just to secure marks where u can if u have done a few rounds of step 3 and still are grappling with those ones, then worst case guess those ones.

yeh thats a smarter approach, thanks for pointing that out

 

[cs123]

Hey do you guys flag the complete the statement/complete the series questions and do them only when you finish all the Set A/b/ questions?

Since with the Set A/B questions once you figure out the pattern you've got 4 questions finished compared to just 1 question finished when you figure out the pattern for complete the... questions

I’ll flag the ones also I think I’ve only gotten the primary pattern for (and not the secondary). after I’ve reached the end of the AR section, I’ll go over the flagged ones one or (rarely) two more times then run out of time.

Hi, would you say it's more efficient to flag and return later, or to completely skip the set and leave them incomplete? It's just that clicking an answer and then flagging five q in a row is somewhat time consuming. Thx!

 

[cs123]

In a q set where SetA has total of even sides and SetB has total of odd sides, for ex., does a line counts as one side length? I have heard that it may not count. What have people regarded lines as in their counting? Appreciate the help!

 

[swagatronium]

Hi, would you say it's more efficient to flag and return later, or to completely skip the set and leave them incomplete? It's just that clicking an answer and then flagging five q in a row is somewhat time consuming. Thx!

O yeah, for me, it depends on whether I've attempted the questions or not.

If I think I've got the primary pattern but am missing the secondary pattern (or vice versa), I'll attempt all the questions and alt+f all the Q's in the set.

If I've not attempted any of the Q's because I haven't got the slightest clue, I'll just leave the Q's all blank so I can alt+i at the end.

That way, I'll do all the Q's I'm confident with and reach the end of the subtest, then I can do all the incomplete questions (that I haven't attempted at all yet), followed by the flagged questions (which I already have attempted, just not sure about).

You could try leaving them all incomplete if you think it eats up some of your time. Though, there's a disadvantage to doing that instead of flagging them all, since once you've put in an answer, you can't go alt+p to those questions. So, if it takes you like 2-3Q's in a set to fully get the pattern, you may have to reach the end of the subtest and then loop back to the same question set to fix those ones. Hope that helps

 

[swagatronium]

In a q set where SetA has total of even sides and SetB has total of odd sides, for ex., does a line counts as one side length? I have heard that it may not count. What have people regarded lines as in their counting? Appreciate the help!

Do you mean a single line itself that's not apart of the shape?

From what you're saying, sounds like the question you're talking about has shapes, yeah? In the case, I think the even and odd pattern is related to the number of sides on the shape itself (i.e. hexagon- 6, triangle - 3), instead of including lines around the shape.

Can't really say for sure though, since I think it can depend on the question itself. For instance, you could have a question that relates to the number of curved and straight sides in each set like:

Set A - even # curved lines, odd # straight lines
Set B - odd # curved lines, even # straight lines

In that case, they could have external lines that do then count to the total "side" count.

I might've not properly understood what you said though, so just lmk if I answered that properly

 

[cs123]

Do you mean a single line itself that's not apart of the shape?

From what you're saying, sounds like the question you're talking about has shapes, yeah? In the case, I think the even and odd pattern is related to the number of sides on the shape itself (i.e. hexagon- 6, triangle - 3), instead of including lines around the shape.

Can't really say for sure though, since I think it can depend on the question itself. For instance, you could have a question that relates to the number of curved and straight sides in each set like:

Set A - even # curved lines, odd # straight lines
Set B - odd # curved lines, even # straight lines

In that case, they could have external lines that do then count to the total "side" count.

I might've not properly understood what you said though, so just lmk if I answered that properly

Yeah, sorrry should have worded that clearer.

I mean when there's a jumble of shapes in the box, there's sometimes 1D lines scattered around the place, and I;m just not sure whether to count them as a side when calculating the total no of sides.

For the pattern you mentioned (with straight and curved lines) I don't think there's much ambiguity there ie defs count the scattered lines that don't form a shape.

Thanks for the explanation tho

 

[ucatboy]

Yeah, sorrry should have worded that clearer.

I mean when there's a jumble of shapes in the box, there's sometimes 1D lines scattered around the place, and I;m just not sure whether to count them as a side when calculating the total no of sides.

For the pattern you mentioned (with straight and curved lines) I don't think there's much ambiguity there ie defs count the scattered lines that don't form a shape.

Thanks for the explanation tho

This doesn't exactly answer your question but honestly if the boxes contain both shapes and a jumble of lines and other weird elements, the likelihood of the pattern being something to do with total number of sides is reduced. Instead, patterns are more likely to depend on position, total elements (not sides), colour or maybe intersections between lines.

 

[cs123]

Can someone tell me the rule for this? Been staring at it for a long time but can't find it.

Did someone identify a rule in the end?

 

[medpharmed]

Hi, I have a question about patterns with odd and even numbers of things. I'm a bit unsure of when to include zero. If set A for example has an even number of intersections but none of the 6 boxes given have 0 intersections, would a test shape with 0 intersections go in set A? Or would we assume that part of the pattern is that there is at least 1 intersection?

 

[kabir1803]

Hi, I have a question about patterns with odd and even numbers of things. I'm a bit unsure of when to include zero. If set A for example has an even number of intersections but none of the 6 boxes given have 0 intersections, would a test shape with 0 intersections go in set A? Or would we assume that part of the pattern is that there is at least 1 intersection?

I would say that 0 would in such a case be neither.

 

[medpharmed]

I would say that 0 would in such a case be neither.

Okay thanks!

 

[rubbishbear]

I'm struggling the most in this section. Shapes just make zero sense to me. Would appreciate some tips for AR! Thank you!

 

[Fili]

I'm struggling the most in this section. Shapes just make zero sense to me. Would appreciate some tips for AR! Thank you!

AR, out of all the sections, is one that you can actually improve the most on because you can just grind your way through it.

Imho just do a lot more questions and read on tips for certain question types if you don't understand it. After that, do some more questions.