BREAKING: Target Test Prep releases Brand New 2026 On Demand GMAT prep course

Redeem

Probability - Roll of the die

This topic has expert replies
Post new topic Post Reply
Previous Topic Next Topic
Junior | Next Rank: 30 Posts
Posts: 27
Joined: Sun Oct 16, 2011 2:34 pm
Thanked: 2 times

Probability - Roll of the die

by dellaboemia » Thu Dec 01, 2011 12:58 pm
A six-sided die with faces numbered one through six is rolled three times. What is the probability that the face with the number 6 on it will NOT be facing upwards on all three rolls?

The correct answer is 1 minus the probability of a 6 facing up on each role. 1 minus 1/6 x 1/6 x 1/6. which comes to 1 minus 1/216 or 215/216.

My first attempt at solving this problem was to take the probability of 1 thru 5 facing up 1/5 and multiplying that by the number of rolls to get 125/216.

My question is this. Why was my approach incorrect?
Top
Source: — Problem Solving |
Master | Next Rank: 500 Posts
Posts: 385
Joined: Fri Sep 23, 2011 9:02 pm
Thanked: 62 times
Followed by:6 members

by user123321 » Thu Dec 01, 2011 1:59 pm
dellaboemia wrote:A six-sided die with faces numbered one through six is rolled three times. What is the probability that the face with the number 6 on it will NOT be facing upwards on all three rolls?

The correct answer is 1 minus the probability of a 6 facing up on each role. 1 minus 1/6 x 1/6 x 1/6. which comes to 1 minus 1/216 or 215/216.

My first attempt at solving this problem was to take the probability of 1 thru 5 facing up 1/5 and multiplying that by the number of rolls to get 125/216.

My question is this. Why was my approach incorrect?
125/216 is correct, when the question says to find...when 6 doesn't face upward in any of the three rolls.
Since they asked to find when 6 doesn't face upward in all of the three rolls, we have 1-1/216

hope this helps.

user123321
Just started my preparation :D
Want to do it right the first time.
Top
Master | Next Rank: 500 Posts
Posts: 176
Joined: Thu Sep 22, 2011 5:32 am
Thanked: 5 times

by vishal.pathak » Thu Dec 01, 2011 2:01 pm
dellaboemia wrote:A six-sided die with faces numbered one through six is rolled three times. What is the probability that the face with the number 6 on it will NOT be facing upwards on all three rolls?

The correct answer is 1 minus the probability of a 6 facing up on each role. 1 minus 1/6 x 1/6 x 1/6. which comes to 1 minus 1/216 or 215/216.

My first attempt at solving this problem was to take the probability of 1 thru 5 facing up 1/5 and multiplying that by the number of rolls to get 125/216.

My question is this. Why was my approach incorrect?
while doing 5*5*5, you are missing the cases where one of the dice has the face of 6 pointing downwards. You are also missing the cases where 2 dice have 6-face pointing downward while the 3rd one has it facing upward

consider one of the dice with the 6-face pointing downward. The other 2 dice can be arranged in 6*5 ways such that all the 3 dice do not have their 6 face pointing downward. 5 in 6*5 is because, i am excluding the condition in which the final dice has its 6-face pointing downward

Now since we have 3 dice, so the number of possible combination in which atleast 1 and atmost 2 dice have their 6-face pointing downward = 3*6*5 = 90

This number is the difference between your answer and the actual answer

Regards,
Vishal
Last edited by vishal.pathak on Thu Dec 01, 2011 2:05 pm, edited 2 times in total.
Top
Master | Next Rank: 500 Posts
Posts: 218
Joined: Wed Nov 23, 2011 8:05 pm
Thanked: 26 times
Followed by:4 members

by chieftang » Thu Dec 01, 2011 2:03 pm
dellaboemia wrote:A six-sided die with faces numbered one through six is rolled three times. What is the probability that the face with the number 6 on it will NOT be facing upwards on all three rolls?

The correct answer is 1 minus the probability of a 6 facing up on each role. 1 minus 1/6 x 1/6 x 1/6. which comes to 1 minus 1/216 or 215/216.

My first attempt at solving this problem was to take the probability of 1 thru 5 facing up 1/5 and multiplying that by the number of rolls to get 125/216.

My question is this. Why was my approach incorrect?



5/6 + (1/6)*(5/6) + (1/6)*(1/6)*(5/6)
Top
Legendary Member
Posts: 1084
Joined: Fri Apr 15, 2011 2:33 pm
Thanked: 158 times
Followed by:21 members

by pemdas » Thu Dec 01, 2011 2:08 pm
The technique is wrong because the three events whose probabilities got multiplied are not mutually exclusive. Within one set you may have one 6 and 1 through 5. Within two events you may have more than one 6, and meaning of the complimentary events {event A=Pr(A), not event A=1-Pr(A)} to which you referred here {1-1/6=5/6} is true for one event where the mutual exclusiveness is observed. When you have two or more events the mutual exclusive principle for the complementary events is not observed, hence you consider the event not A with the less probability than it's actually. You keep counting and relying on 5/6 probability rule for the dice not facing with 6, whereas the dice will face with 6 more than once now and other numbers will face with numbers 1-5 more than five times now, they are replicated many-many times for two experiments (events). Try to understand that complementary events and their probabilities are centered around the principle of mutual exclusiveness of outcomes within one experiment (one event)
dellaboemia wrote: My first attempt at solving this problem was to take the probability of 1 thru 5 facing up 1/5 and multiplying that by the number of rolls to get 125/216.

My question is this. Why was my approach incorrect?
Success doesn't come overnight!
Top
Legendary Member
Posts: 1084
Joined: Fri Apr 15, 2011 2:33 pm
Thanked: 158 times
Followed by:21 members

by pemdas » Thu Dec 01, 2011 2:12 pm
sorry, but this is not correct; in any of three attempts=all three attempts. If you were to say in single attempts taken as separate experiments, then you're on. But here it's not so as you explain
user123321 wrote: 125/216 is correct, when the question says to find...when 6 doesn't face upward in any of the three rolls.
Since they asked to find when 6 doesn't face upward in all of the three rolls, we have 1-1/216

hope this helps.

user123321
Success doesn't come overnight!
Top
Junior | Next Rank: 30 Posts
Posts: 27
Joined: Sun Oct 16, 2011 2:34 pm
Thanked: 2 times

by dellaboemia » Thu Dec 01, 2011 6:07 pm
Thanks all for the clarification! Clearly, my probability and statistics skills need refreshing. I've seen posts addressing similar questions / coin flip, etc... I'll be researching this area further for strategies. Thanks again. Feeling quite supported!
Top
Legendary Member
Posts: 966
Joined: Sat Jan 02, 2010 8:06 am
Thanked: 230 times
Followed by:21 members

by shankar.ashwin » Thu Dec 01, 2011 7:41 pm
A few problems are easier if you just list down a few possibilities and see a pattern.

Here Total outcomes = 6*6*6 = 216.

Unfavorable outcome = 1 (6,6,6)

All other outcomes are favorable; for eg; (1,1,1) (5,6,6) (2,3,4) etc. You have 215 such outcomes (everything except (6,6,6) )

So required probability = N(Favorable)/N(total) = 215/216
Last edited by shankar.ashwin on Fri Dec 02, 2011 12:55 am, edited 1 time in total.
Top
Legendary Member
Posts: 1084
Joined: Fri Apr 15, 2011 2:33 pm
Thanked: 158 times
Followed by:21 members

by pemdas » Fri Dec 02, 2011 12:00 am
@shankar: "6" doesn't appear in any of the three rolls and "6" doesn't face in all three rolls is the same. Pr(not 6)=5/6 is true for only one experiment. For successive experiments (call it replications in stats) we have to adjourn the reference to complementary event (not 6) and count only individual probabilities. The events will not longer be mutually exclusive for the experiments replicated after the first experiment.
Success doesn't come overnight!
Top
Legendary Member
Posts: 966
Joined: Sat Jan 02, 2010 8:06 am
Thanked: 230 times
Followed by:21 members

by shankar.ashwin » Fri Dec 02, 2011 1:00 am
Ah get your point.

1st roll P(not 6) = 5/6

2nd roll P(not 6) = (5/6)*(5/6) - Prob that 6 does not appear in 1st and then prob of 6 not appearing in second

3rd roll P(not 6) = (5/6)*(5/6)*(5/6)

Together it would be 5^6/6^6 i guess. Again not sure :)
pemdas wrote:@shankar: "6" doesn't appear in any of the three rolls and "6" doesn't face in all three rolls is the same. Pr(not 6)=5/6 is true for only one experiment. For successive experiments (call it replications in stats) we have to adjourn the reference to complementary event (not 6) and count only individual probabilities. The events will not longer be mutually exclusive for the experiments replicated after the first experiment.
Top
Post new topic Post Reply
• Page 1 of 1