Is one estimator quietly pricing higher than the other?
Eleven jobs priced by each of two estimators, in dollars per square meter. The gap is $122.73 and it is nowhere near noise, which on a 2,000 square meter package puts a quarter of a million dollars on who happened to pick up the file.
Construction Intermediate Statistics free
After you install, this is the model to open.
Do Our Two Estimators Price the Same Job the Same Way?
- In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
- Click Start from a template and put that name in the search box.
- Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.
The answer
- The gap
- $122.73 a square meter, Ravi over Sian, on the same drawings
- Real or luck
- p < 0.001 two-tail 0.0000017, t = 6.68, far below alpha 0.05
- The gap the data supports
- $84 to $161 95% interval, nowhere near zero
- Same eleven jobs
- 11 plans priced independently by both estimators
Eleven jobs priced by Ravi and eleven different jobs priced by Sian, all in dollars per square meter. Ravi averages $1,393.64 and Sian averages $1,270.91, a gap of $122.73 a square meter. Click Run: t comes back at 6.68 on 20 degrees of freedom with a two-tail p-value of 0.0000017. Two estimators working to the same rate book would produce a gap this wide about twice in a million comparisons, so this is a systematic difference in how the two of them price, not a run of luck.
The sheet turns it into the number that matters: on a 2,000 square meter package, which estimator opened the file is worth $245,455. That is bigger than most of the contingencies these bids carry. Note carefully what this test can and cannot be. These are twenty-two different jobs, so the comparison is between two independent groups, and part of the gap could be that Ravi happened to get the harder work.
The two scope columns are on the sheet so you can check the mix before you believe the number. If you want to remove the possibility altogether, hand both estimators the same eight jobs, put the two prices side by side and run t-Test: Paired instead; that version answers the question with no scope confound at all and is the run to do before anyone is asked to explain themselves.
The equal-variances assumption holds here: F-Test: Two-Sample for Variances on the same two ranges returns F of 1.65 with a one-tail p-value of 0.220, on sample variances of 2,315 and 1,399. Run that check on your own data and switch to the Welch version, t-Test: Unequal Variances, if it fails. What the test cannot tell you is which estimator is right.
A consistent gap of about nine percent between two people is equally consistent with one of them being careful and the other being optimistic, and the only way to settle that is to compare both against what the jobs actually cost to build. To use your own data, paste one estimator per column and widen both ranges.
The model
It arrives on a tab called Template: Two Estimators, One Rate, carrying these columns:
- Priced by Ravi ($/m2)
- Job
- Priced by Sian ($/m2)
- Scope (Ravi)
- Scope (Sian)
with the model computed beside the data:
| Difference in average rate ($/m2) | 122.7 |
| On a 2,000 m2 package ($) | 245,454.5 |
Once it is in your sheet
- The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
- Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
- Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Next question
- What will a tonne of steel cost you next month?Where Is This Material Price Going?
- How much contingency does the job need before the budget holds nine times in ten?What contingency gives 90% odds of staying under budget?
- What happens to the reserve when costs rise together?Will the Contingency Last?
- Which finish date can you put in the contract?The P80 Construction Schedule
- What are the odds the rain costs you the date?Will the Weather Blow the Deadline?
- What is finishing on time actually worth?What Do Late Days Really Cost?
Every model like this one, and the method behind them: Statistics in Google Sheets.