A consulting firm tracks performance metrics for eight regional office

Question

A consulting firm tracks performance metrics for eight regional offices, four in each of two geographic regions (East and West). The following table contains these metrics for calendar year 2019. This firm hires consultants for its regional offices on a calendar-year basis, so the number of consultants employed at each office listed did not change at any point during 2019.

2019 Regional Office Performance Data

Office  
  Region  
  Consultants  
  Projects   Completed  
  Avg Project   Revenue ($k)  
  Total   Costs ($M)  
  Client   Satisfaction (%)   
 
  A 
 West 
 42 
 115 
 86 
 7.9 
 91 
 
  B 
 East 
 35 
 104 
 93 
 7.4 
 88 
 
  C 
 West 
 51 
 135 
 79 
 9.8 
 94 
 
  D 
 East 
 28 
 96 
 102 
 7.1 
 85 
 
  E 
 East 
 47 
 122 
 89 
 8.6 
 92 
 
  F 
 East 
 39 
 110 
 90 
 8.1 
 90 
 
  G 
 West 
 33 
 89 
 95 
 6.2 
 87 
 
  H 
 West 
 44 
 120 
 84 
 8.5 
 93

Each of the following statements, if true, predicts the existence of a correlation between two of the metrics in the table. For each statement, select  Consistent  if the predicted correlation holds for the data in the table. Otherwise, select  Inconsistent .

RonPurewal · Accepted Answer

A consulting firm tracks performance metrics for eight regional offices, four in each of two geographic regions (East and West). The following table contains these metrics for calendar year 2019. This firm hires consultants for its regional offices on a calendar-year basis, so the number of consultants employed at each office listed did not change at any point during 2019.

2019 Regional Office Performance Data

Office	Region	Consultants	Projects Completed	Avg Project Revenue ($k)	Total Costs ($M)	Client Satisfaction (%)
A	West	42	115	86	7.9	91
B	East	35	104	93	7.4	88
C	West	51	135	79	9.8	94
D	East	28	96	102	7.1	85
E	East	47	122	89	8.6	92
F	East	39	110	90	8.1	90
G	West	33	89	95	6.2	87
H	West	44	120	84	8.5	93

Each of the following statements, if true, predicts the existence of a correlation between two of the metrics in the table. For each statement, select Consistent if the predicted correlation holds for the data in the table. Otherwise, select Inconsistent.

Part I: “According to the psychological phenomenon of attachment to sunk costs as well as the more general principle of psychological commitment and consistency, people’s self-image and emotions become more and more invested in achieving a goal as total expenditures in pursuit of that goal increase. Therefore, as total project expenditures increase, clients will become more and more strongly biased toward reporting complete satisfaction with the consultants’ work and the project outcomes.”

In this statement there’s no possible confusion about the correlation being asserted: “as total project expenditures increase, clients will become more and more strongly biased toward reporting complete satisfaction.” So, we need to look for a positive correlation between Total Costs and Client Satisfaction Rate.

We can sort by either one of these two variables (for the same reason why it doesn’t matter which variable goes on the x-axis of a scatterplot, if we’re looking only for correlation). If we sort by Total Costs, the two columns of interest are as follows:

The positive correlation is obvious. (Note how all the percentages in the 80s come first, followed by all the ones in the 90s.) So this statement is Consistent with the table.

Part II: “The more consultants who work in a given office, the broader the range of their individual specialties and areas of expertise—and thus the greater the likelihood that each of the office’s projects will be staffed with consultants with the necessary competence and background to perform the work to the complete satisfaction of the client.”

The correlation ultimately being predicted here is between Number of Consultants (“the more consultants who work in a given office...”) and Client Satisfaction Rating (“...to the complete satisfaction of the client”). Sort by either of these and then check for the necessary (positive)
correlation.

If we sort by Number of Consultants, the results are as follows:

Once again, the positive correlation is obvious. (Were it not for the 93% and 92% values coming out of order, the correlation here would be perfect.) This statement is also Consistent with the table.

Part III: “For some of the company’s projects, the client’s goals include intangible objectives, such as development of the client’s cross-cultural insight and leadership skills, in addition to revenue generation. Although these projects typically generate less revenue than those strictly focused on cash flow and profitability gains, they almost invariably receive perfect client satisfaction ratings.”

We have no information whatsoever about project goals, so the relevant correlation is between the other two metrics mentioned here: Average Project Revenue (“these projects typically generate less revenue”) and “client satisfaction ratings”. So, we need to sort the rows by one of these variables (once again it doesn’t particularly matter which one), and then check to see whether the resulting columns are ANTI-correlated (as one goes up, the other goes down).

Here’s what we get from sorting by Average Project Revenue:

Just like last time, the desired correlation is as close as possible to perfect without actually being perfect (marred only by the 92% and 91% figures coming out of order). So this statement, too, is very much Consistent with the table.

Show Spoiler

Attachment:

GMAT-Club-Forum-xgwcceuy.png [ 84.49 KiB | Viewed 996 times ]

Attachment:

GMAT-Club-Forum-4gbafy7c.png [ 22.6 KiB | Viewed 966 times ]

Attachment:

GMAT-Club-Forum-22gmb6vy.png [ 21.52 KiB | Viewed 950 times ]

Attachment:

GMAT-Club-Forum-qa6qaofm.png [ 23.9 KiB | Viewed 936 times ]

Edskore · Answer

This is a Graphs and Tables correlation verification question — one of my favourite DI formats because it tests careful data reading rather than computation. The key skill: don't trust the story the statement tells you. Go straight to the numbers and check.

The question gives you a table of 8 offices and three statements, each predicting a correlation. You have to verify whether each correlation actually holds in the data.

The approach: For each statement, identify the two metrics being correlated, then check whether the predicted direction (positive or negative) is consistent across the table.

Statement 1 — Consistent
Prediction: Higher total costs → higher client satisfaction (sunk cost bias argument).

Sorting offices by total costs ascending: G (6.2M, 87%), D (7.1M, 85%), B (7.4M, 88%), A (7.9M, 91%), F (8.1M, 90%), H (8.5M, 93%), E (8.6M, 92%), C (9.8M, 94%). The trend is strongly upward. Out of all 28 pairwise comparisons, 25 are concordant (both variables move in the same direction) vs. only 3 discordant. Consistent.

Statement 2 — Consistent
Prediction: More consultants → higher client satisfaction (broader expertise).

Sorting by consultants: D (28, 85%), G (33, 87%), B (35, 88%), F (39, 90%), A (42, 91%), H (44, 93%), E (47, 92%), C (51, 94%). Nearly perfect — only one exception (H to E: 44 to 47 consultants, but satisfaction dips 93 to 92). 27 of 28 pairs are concordant. Consistent.

Statement 3 — Inconsistent
Prediction: Lower average project revenue → higher client satisfaction (intangible-goal projects generate less revenue but higher satisfaction). This predicts a negative correlation between revenue and satisfaction.

But check the data: sorting by avg revenue ascending — C (79k, 94%), H (84k, 93%), A (86k, 91%), E (89k, 92%), F (90k, 90%), B (93k, 88%), G (95k, 87%), D (102k, 85%). Satisfaction actually decreases as revenue decreases — they move together in the same direction, not opposite directions. 27 of 28 pairs run against the predicted negative correlation. Inconsistent.

Common trap: Reading the sunk-cost argument in Statement 1 and dismissing it as "that can't be real data" — on GMAT, you verify against the table, not against your intuition about causation. Similarly, Statement 3 sounds plausible as a business theory, but the data flatly contradicts it.

Takeaway: In G&T correlation questions, translate each statement into a predicted direction (positive or negative), sort the data by one variable, then check whether the other variable moves the way the statement predicts.

(Kavya | 725 on GMAT Focus Edition)

chetan2u · Answer

A table is given along with statements in the table that require to be checked for their consistency.

Of course we require to check the table but initially we would require to test statements as the metrics that have to be compared are not given in a very straight fashion. So we look for comparative words , increase/decrease or more/less'.

1.According to the psychological phenomenon of attachment to sunk costs as well as the more general principle of psychological commitment and consistency, people’s self-image and emotions become more and more invested in achieving a goal as total expenditures in pursuit of that goal increase. Therefore, as  total project expenditures  increase  ,  clients will become  more and more  strongly biased toward reporting complete satisfaction  with the consultants’ work and the project outcomes.
 Here the portions in red and blue are compared and both move in the same direction as shown by the underlined portion. 
So we require to check whether total expdr and client's satisfaction move in same direction. When we sort the table as per increasing expdr, we find that even satisfaction columm is sorted increasingly.
So we can say the statement is  consistent.

2. The  more consultants  who work in a given office, the broader the range of their individual specialties and areas of expertise—and thus the greater the likelihood that each of the office’s projects will be staffed with consultants with the necessary competence and background to perform the work to the  complete satisfaction of the client.

Here the portions in red and blue are compared and both move in the same direction as shown by the underlined portion. 
So we require to check whether more number of consultant and client's satisfaction move in same direction. When we sort the table as per increasing number of consultants, we find that even satisfaction column is sorted increasingly.
So we can say the statement is  consistent.

3. For some of the company’s projects, the client’s goals include intangible objectives, such as development of the client’s cross-cultural insight and leadership skills, in addition to revenue generation. Although these projects typically generate  less revenue  than those strictly focused on cash flow and profitability gains, they  almost invariably receive perfect client satisfaction ratings.

Here the portions in red and blue are compared and both move in the opposite direction as shown by the underlined portion. Less revenue leads to complete satisfaction. 
So we require to check whether less revenue and client's satisfaction move in opposite direction. When we sort the table as per decreasing revenue, we find that satisfaction column gets sorted in decreasing order.
So we can say the statement is  consistent.

All statements are consistent.

chetan2u · Answer

You may have to relook at the third statement.

RonPurewal · Answer

Look over those figures again... You've sorted them in increasing order by revenue per project (the first number in each pair), so, the nature of the correlation (if any) will be determined by whether the  second  values primarily increase, decrease, or neither. (The GMAT will not test you on recognizing weak patterns. If you need to identify a correlation to solve a problem, then that correlation will be either perfect or near-perfect.)

Let's pull those numbers out of the ordered pairs so we can more easily see whether they march up or down:
94%
93%
91%
92%
90%
88%
87%
85%

This series of numbers is in almost perfect decreasing order. (Only the 91% and 92% keep the decreasing order from being completely perfect.) So there's clearly a  strong anti-correlation  (NEGATIVE correlation) between these two statistics—precisely what you need to establish  consistency  with the third statement, as you correctly laid out above.

Not sure how you're seeing a positive correlation at work here; if you still think there's a positive correlation, I wouldn't mind seeing the numerical reasoning behind that.

.

A correlation is a property of the  whole dataset —not a property of random pairings of 2 out of the N given datapoints. Just make an ordered list of ALL the datapoints you're treating—or perhaps a scatterplot—and then look for an OBVIOUS, NEAR-PERFECT (or even actually perfect) series of positive or negative increments in the y-variable as you advance the x-variable.

Remember, the GMAT will  NEVER  require you to recognize weak or "subtle" correlations;  any correlation you need to recognize will be apparent at a glance  at a list or plot of THE WHOLE DATASET.

.

(Counting pairwise outcomes isn't a valid way to determine correlations in general, anws. If you're given a correlation as strong as the ones GMAC asks you to identify, then certainly a majority of the pairs will correlate in the same direction as the overall dataset does—but why do all that extra work, in exchange for a LESS direct view of the correlation as a whole?
More generally—outside of the standardized-testing world—you can definitely have a dataset whose overall correlation runs against a majority of the possible pairwise selections of points: e.g., a scatterplot that shows a positive correlation overall but in which there are more anti-correlated pairs than positively correlated pairs. Just imagine a dataset framed by 3 points with huge, unmistakable steps upward in both variables.... interspersed with 5-6 intermediate datapoints with VERY SLIGHT steps down in 'y' as 'x' goes up. The 3 points creating the strong frame will dominate the statistical analysis and give a distinct positive correlation overall, even though you'll have more anti-correlated pairs if you insert enough ever-so-slightly-decreasing sets of points in the middle.)

MacT750 · Answer

just was confused about perfect client rating. perfect rating means a100% no?

RonPurewal · Answer

If you're referring to the third part, note that only   some  —NOT all—projects at the offices fitting this description have the characteristics enumerated (= include intangible objectives that don't yield revenue, but nearly always garner perfect satisfaction ratings).

This sub-group of projects will push the overall average revenue per project DOWN and the overall satisfaction rating UP for those offices, relative to the others without such projects. They're not going to raise the overall satisfaction rating for the entire year's work to 100% at any office, though, because they are not 100% of the projects undertaken by any office.

Bunuel · Answer

Similar questions based on the same table:
https://gmatclub.com/forum/a-consulting ... 55422.html
https://gmatclub.com/forum/a-consulting ... 55421.html
https://gmatclub.com/forum/a-consulting ... 55420.html

egmat · Answer

Hi MacT750,

Great question! This is a  common point of confusion  on these correlation-based TA questions.

You're right that 'perfect client satisfaction ratings' would normally mean  100% . But here's the  key insight : S3 is NOT asking you to verify whether any office literally has  100%  satisfaction. The question asks whether the PREDICTED CORRELATION holds in the data.

S3's logic boils down to this: some projects focus on intangible goals rather than revenue, so they generate less revenue but get higher satisfaction. The predicted  correlation  is  therefore : lower average revenue should correlate with higher satisfaction.

Let's check by sorting Avg Project Revenue from highest to lowest and looking at satisfaction:

- D:  $102k  revenue →  85%  satisfaction
- G:  $95k  →  87% 
- B:  $93k  →  88% 
- F:  $90k  →  90% 
- E:  $89k  →  92% 
- A:  $86k  →  91% 
- H:  $84k  →  93% 
- C:  $79k  →  94%

Key Insight:  As average revenue goes DOWN, satisfaction goes UP — almost perfectly!  That  negative correlation  is exactly what S3 predicts, so it's  Consistent .

The 'perfect ratings' language is part of the causal story explaining WHY this  correlation  might exist. It's the mechanism behind the prediction. But you're only asked to verify whether the  correlation  itself shows up in the data — not whether individual details of the explanation (like literal  100%  scores) appear in the table.

Trap:  Don't get distracted by the literal wording of the causal mechanism.  Focus on the direction of the correlation being predicted.

Answer: Consistent, Consistent, Consistent

A consulting firm tracks performance metrics for eight regional office

Prep Toolkit

Top 5 GMAT Debrief Videos

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards

Consistent	Inconsistent
		According to the psychological phenomenon of attachment to sunk costs as well as the more general principle of psychological commitment and consistency, people’s self-image and emotions become more and more invested in achieving a goal as total expenditures in pursuit of that goal increase. Therefore, as total project expenditures increase, clients will become more and more strongly biased toward reporting complete satisfaction with the consultants’ work and the project outcomes.
		The more consultants who work in a given office, the broader the range of their individual specialties and areas of expertise—and thus the greater the likelihood that each of the office’s projects will be staffed with consultants with the necessary competence and background to perform the work to the complete satisfaction of the client.
		For some of the company’s projects, the client’s goals include intangible objectives, such as development of the client’s cross-cultural insight and leadership skills, in addition to revenue generation. Although these projects typically generate less revenue than those strictly focused on cash flow and profitability gains, they almost invariably receive perfect client satisfaction ratings.

GMAT Graphs and Tables (G&T) Questions

A consulting firm tracks performance metrics for eight regional office

Prep Toolkit

615 to 715 in 15 Days (Ria's Strategies)

More than Quant/Verbal Abilities: Speed vs. Accuracy

745 with Only Self-Prep and GMAT Club

From 555 to 765: 210-point Improvement

Perfect 805 Score Debrief by Julia

My Rewards