Chapter 20 Quiz — Labor, Productivity, and the Construction Workforce

Twenty questions. Answer each before opening the explanation — retrieval is the point, and looking first feels productive while teaching you nothing. Scoring guide at the end.

Multiple Choice (10)

Q1. In the identity unit cost = quantity × productivity × rate, which term does a project team have the most control over after Notice to Proceed?

A. Quantity B. Productivity C. Rate D. All three roughly equally

Answer

B — Productivity. Quantity is fixed by the drawings and changes only through change orders and waste. Rate is set by the labor market, the wage scale or determination, your insurance carrier, and your accountant — you can nudge it through crew mix, but not much. Productivity is set by sequence, access, staging, crew size, hours, and supervision, all of which are your decisions. This asymmetry is why an estimate can be right and a job can still lose money.

Q2. An activity budgeted at 2,000 MH is 55% complete with 1,250 MH charged. The productivity factor is:

A. 0.63 B. 0.88 C. 1.14 D. 1.60

Answer

B — 0.88. Earned MH = 0.55 × 2,000 = 1,100. PF = earned ÷ actual = 1,100 ÷ 1,250 = 0.88. You are getting 88 cents of budgeted work out of every hour purchased. Forecast at completion = 2,000 ÷ 0.88 = 2,273 MH, about 273 hours over.

Q3. Why does a lower-wage classification typically carry a higher burden multiplier than a higher-wage classification in the same trade?

A. Lower-wage workers have higher workers' compensation rates B. Several burden components are flat dollars per person rather than percentages of wage C. Apprentices work more overtime D. FICA is regressive above the wage base

Answer

B. Health and welfare, small tools, and overhead allocation cost roughly the same per person regardless of wage. Divided into a smaller base wage, those flat dollars produce a larger multiplier. In §20.5 the apprentice at $23.80 carries a 1.71 multiplier against the foreman's 1.54. The practical consequence: you cannot price a mixed crew by applying one multiplier to one average wage — you must price each seat and blend.

Q4. The correctly computed cost of one overtime hour is closest to:

A. 1.5 × the fully burdened straight-time rate B. 1.5 × the base wage C. The burdened straight-time rate plus the half-time premium on the base wage, grossed up for statutory burdens D. The burdened straight-time rate plus 50% of all burden components

Answer

C. The half-time premium applies to the base wage, and statutory burdens (FICA, unemployment, workers' compensation, general liability) apply to the premium because it is wages. Fixed components — health and welfare, tools, overhead allocation — do not repeat with the extra hour. On Kestrel's carpenter: $54.12 + ($17.00 × 1.2128) = $74.74, about 1.38× straight time. Pricing at 1.5 × burdened overstates each overtime hour by roughly $6.44.

Q5. In a work-sampling study, the most useful finding is usually:

A. The direct-work percentage compared to an industry benchmark B. Which non-direct category is largest on your job C. The idle percentage D. The ratio of direct work to preparatory work

Answer

B. Direct-work percentages vary enormously by trade, phase, and study definition, so benchmarking yours against a published figure tells you almost nothing. The largest non-direct category is diagnostic: travel points at logistics and staging, waiting points at constraint removal and sequencing, material handling points at procurement and delivery, preparatory work points at coordination and information flow. Idle is almost never the largest, which is itself the lesson.

Q6. On a prevailing-wage project, a worker performs carpenter work for six hours and laborer work for two hours in the same day. The correct treatment is:

A. Pay the carpenter rate for all eight hours B. Pay the laborer rate for all eight hours C. Pay a blended average of the two rates D. Split the hours by classification, supported by contemporaneous records

Answer

D. Hours must be split by the classification of work actually performed, and the split must be supported by records made at the time. Paying the higher rate throughout (A) is compliant but expensive; a blend (C) is not a recognized method; paying the lower rate throughout (B) is an underpayment. The record is the hard part, and it is a field responsibility, not a payroll-department one.

Q7. A crew of 9 produces 780 SFCA/day at a cost of $4,442/day. Adding a tenth person raises production to 840 SFCA/day at $4,983/day. The marginal cost of the added production is:

A. $5.93 per SFCA B. $5.70 per SFCA C. $9.02 per SFCA D. $54.12 per hour

Answer

C — $9.02. Extra cost $541 ÷ extra production 60 SFCA = $9.02 per SFCA, against an average unit cost at the optimum of $5.70. You are buying production at roughly 58 percent above budget. That can still be the right call — if you are buying a date on the critical path — but you should know that is what you are doing, in writing. All ten people will look busy either way, which is exactly why you measure instead of eyeballing.

Q8. Which of the following best explains why Kestrel runs union on public work and open shop on most private work?

A. Public owners require union labor B. Prevailing-wage determinations largely erase the open-shop rate advantage on public jobs C. Union crews are more productive on public work D. Open-shop contractors cannot obtain payment and performance bonds

Answer

B. The wage determination sets the package everyone must pay, so the largest component of the open-shop advantage disappears while the union model's advantages — trained craft on call, shared training cost — remain. A is false as a general statement (some jobs have a PLA, most do not); C is unsupported (the model is not the variable, the training system is); D is simply wrong.

Q9. A subcontractor's manpower loading curve shows 116 workers against a planned 142 in week 32, and the schedule update shows no late activities. The correct interpretation is:

A. The schedule is fine; the curve is a planning artifact B. The subcontractor has found efficiencies C. A duration built on 142 workers is already stretching, and the schedule update has not caught up yet D. The planned curve was overstated at buyout

Answer

C. Durations are built from crew size and unit rate. Staffing 82 percent of plan means the duration is stretching by roughly the inverse, even though no activity has reported late — activities report late only after the fact. The manpower curve leads the schedule update by roughly four to six weeks, which is the difference between making a phone call and writing a delay notice. D is a legitimate alternative hypothesis worth testing, but you test it by asking, not by assuming.

Q10. Which of these is the strongest basis for a lost-productivity claim?

A. A published productivity adjustment factor table applied to bid hours B. A measured-mile comparison of the contractor's own unimpacted and impacted performance on the same work C. A work-sampling study performed during the impacted period D. Testimony from the foreman about how difficult conditions were

Answer

B — the measured mile. It uses the contractor's own performance as its own baseline, which removes the argument that the bid was optimistic. Factor tables (A) are planning heuristics and are routinely dismantled by owners' consultants. A work sample during the impacted period (C) has no baseline to compare against. Testimony (D) has value only as corroboration. See Chapter 33.

True / False — justify each in one line (5)

Q11. An apprentice is always cheaper per unit installed than a journeyman because the apprentice's wage is lower.

Answer

False. Cost per unit is wage divided by output, and an apprentice at 65 percent of scale producing 60 percent of output is more expensive per unit. You staff apprentices for the pipeline, not for the price — a company decision with a company payback, which is why evaluating it one job at a time always produces the wrong answer.

Q12. Compressing a work week (three 12-hour days) and extending one (six 10-hour days) cost the same premium per hour worked.

Answer

False. Three 12-hour days is 36 hours — under 40, so in most jurisdictions no weekly overtime premium applies at all, while six 10-hour days generates 20 premium hours. The caveat matters: some states require daily overtime after 8 or 10 hours regardless of the weekly total, and many labor agreements do the same. Check both before you use the move.

Q13. A general contractor has no reason to care whether a subcontractor classifies its workers as employees or independent contractors, because they are not the GC's employees.

Answer

False. Liability for a lower tier's wage or workers' compensation failure can reach up the chain under many state statutes, an uninsured injury on your site becomes your problem very quickly, and prevailing-wage certified payrolls that do not reflect the people actually working are your compliance exposure too. Prequalify, verify coverage before badging, and keep the right to withhold for documented non-compliance.

Q14. Sustained overtime's efficiency loss is roughly constant from the first week onward.

Answer

False. The loss is progressive — small in week one and substantial by week four or five, as fatigue accumulates, absenteeism rises, pace self-adjusts to the longer schedule, and support functions sized for 40 hours fail to scale. This is why short bursts of overtime are an efficient tool and sustained overtime is a way of spending money to go slower.

Q15. A safety incentive program that pays a bonus for injury-free months reliably improves safety.

Answer

False. It frequently improves reported safety by suppressing reporting — peer pressure on an injured worker, a foreman finding a reason not to record. OSHA's recordkeeping rule includes anti-retaliation provisions, and OSHA has issued guidance that incentive programs discouraging reporting can themselves be a problem. Incentivize leading indicators — near-misses reported, hazards corrected, inspections completed — never lagging ones.

Short Answer (4)

Q16. A cost code goes bad. List, in order, the five things you check — and say why the order matters.

Answer

(1) Is the measurement real? Recount the quantity physically and pull the time cards; wrong inputs make every downstream conclusion wrong, so this is always first. (2) Did the work environment change? Access, sequence, trade stacking, material at the face, information flow. (3) Did the crew change? Size, mix, foreman, overtime, absenteeism, turnover. (4) Is there rework hiding in the code? Check nonconformance and RFI logs. (5) Only then, and rarely, effort. The order matters because investigating in any other sequence anchors you on a conclusion — usually "the crew is slow" — that the data was never going to support.

Q17. Explain the difference between a rate variance and a productivity variance in labor cost, and why reporting only the total is a management failure.

Answer

Productivity variance = (earned MH − actual MH) × budget rate: you used more hours than the work earned. Rate variance = actual MH × (budget rate − actual rate): the hours you used cost more than budgeted, typically from overtime premium or a richer crew mix. Reporting only the total sends the wrong person to fix it — a rate variance is fixed in the office by stopping overtime or changing crew mix; a productivity variance is fixed in the field by clearing access, staging material, and unstacking trades. One number, two different buildings.

Q18. Name four structural barriers that keep women a small minority of the craft workforce, and state the cost of fixing each.

Answer

Any four of: facilities — no separate, clean, secure, well-located toilet (cost: a second sanitary unit and a lock); harassment and hostile-site behavior (cost: an enforced policy, supervisor training, and one credible removal — free, and it costs a supervisor's comfort); PPE that does not fit (cost: buying the right sizes, and it is a safety issue for anyone the standard sizes do not fit); schedule unpredictability (cost: look-ahead discipline you should have anyway); no sponsorship for advancement (cost: a promotion process with named candidates — free); isolation as the only one on a crew (cost: deliberate crew assignment — free). The point of listing the costs is that this is a supply problem with cheap fixes, not a values debate.

Q19. Why is a craft labor shortage a schedule problem before it is a cost problem?

Answer

Because it does not announce itself as a forecastable wage increase. It announces itself as a subcontractor who bid four crews and can staff two. Durations were built from crew size and unit rate; half the assumed crew roughly doubles the duration, and you learn it in week three of a fourteen-week activity. Money can be forecast, priced, and carried in contingency. Absent people cannot be bought at any price — you can only replan around them, spending float you have already committed.

Applied Scenario (1)

Q20. You are the PM on a $31 million job. An activity budgeted at 4,200 MH is 58 percent complete with 2,900 MH charged. Burdened rate $56.00/hr. Your superintendent says the crew is fine and the problem is that the estimate was low. The activity has been performed in an area where two other trades were working for the last three weeks, and the crew went to Saturdays four weeks ago.

Compute the PF, the forecast, and the variance in dollars. Then say whether you accept the superintendent's explanation, and name the two pieces of evidence that would settle it.

Answer

The arithmetic. Earned = 0.58 × 4,200 = 2,436 MH. PF = 2,436 ÷ 2,900 = 0.84. Forecast = 4,200 ÷ 0.84 = 5,000 MH (cross-check: 2,900 ÷ 0.58 = 5,000). Variance = 5,000 − 4,200 = 800 MH = $44,800.

Do you accept the explanation? Not yet — and note that "the estimate was low" is the one explanation that cannot be acted on, which is part of why people reach for it. Two conditions in the scenario are known productivity depressants: three or more trades in one area (a 1.15–1.40 factor band) and four weeks of Saturdays (a fatigue band). Either could produce a 0.84 by itself.

The two pieces of evidence that settle it. (1) The PF trend, week by week, from the start of the activity. If the PF was near 1.00 for the first six weeks and fell after the trades stacked and the Saturdays began, the estimate was fine and the conditions changed. If the PF was 0.84 from week one, the estimate — or the crew's method — is genuinely the issue. This single chart is why weekly measurement is worth the trouble. (2) The crew composition and hours record, compared against what the estimate assumed: headcount, journeyman-to-apprentice ratio, and overtime hours as a percentage of total. A crew that quietly lost two journeymen, or that has been at 33 percent overtime for a month, explains the number without anyone being at fault.

What you do next, regardless. Forecast to completion at 0.84 and tell somebody today — this activity cannot be recovered, only contained. Then compute what containment is worth: lifting the remaining 42 percent from 0.84 to 0.95 takes the forecast from 5,000 MH to about 2,900 + (1,764 ÷ 0.95) = 4,757 MH, roughly 243 hours and $13,600 saved on work you had already lost.

Scoring Guide

70 percent (14/20) is the threshold to proceed. But if you missed Q2, Q4, Q7, or Q20, go back regardless of your total — those four are the arithmetic you will actually use on Monday.