# Answer key · MFG01

## Did sewing and finishing teams meet their recorded productivity targets equally often?

Production managers compare departments to decide where to look first when output falls short. Before doing that, they need to know how a team-day counts as meeting its target and how many team-days each department contributes.

- One record is — one team-day record
- Period — 2015-01-01 to 2015-03-11
- Scope — One garment company; its location is not confirmed in the UCI metadata.

## The table

Target attainment × Department

| Target attainment | Sewing | Finishing |
| --- | ---: | ---: |
| Met or exceeded target | 573 | 302 |
| Below target | 118 | 204 |

## Which statement does this table support?

1. Sewing team-days met their recorded target more often than Finishing team-days.
2. Moving a team from Finishing to Sewing would raise its chance of meeting target.
3. Sewing teams are more productive than Finishing teams.

Choose one statement. Show the counts and denominators that justify it, then describe one limit of the comparison.

## The figures behind the check

- **1. Supported.** 573 of 691 (82.9%) against 302 of 506 (59.7%): 23.2 percentage points higher, in one company’s records.
- **2. Not supported.** That is a causal claim. The departments differ in work, targets and staffing, and nothing here holds those constant; the table records what happened, not what a change would do.
- **3. Not supported.** Each team-day is judged against its own recorded target, and targets may be set differently in each department. The table compares attainment against those targets, not output.

- Sewing: 573 of 691 team-days at or above target (82.9%).
- Finishing: 302 of 506 team-days at or above target (59.7%).
- Sewing supplies 57.7% of all 1,197 team-days, so a pooled attainment rate leans toward it.

Report the gap as an association in one company’s records, not as a department effect. Asking what a change would do needs like-for-like work and target rules.

## Limits

Historical records from one garment company, not an industry sample. Teams recur across dates, so the records are not independent workers. The UCI metadata does not confirm where the company is. Some actual_productivity values exceed one, outside the documented zero-to-one range; they are kept as recorded until the authors clarify. All summaries are descriptive, not causal.

## Source

Data: Productivity Prediction of Garment Employees [Dataset]. (2020). UCI Machine Learning Repository. https://doi.org/10.24432/C51S6D. Licensed under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Crosstabs selected, recoded and aggregated the records into this table.

- [Productivity Prediction of Garment Employees [Dataset]. (2020). UCI Machine Learning Repository. https://doi.org/10.24432/C51S6D](https://archive.ics.uci.edu/dataset/597/productivity+prediction+of+garment+employees)
- [Reproduce it · MFG01](https://www.crosstabs.com/examples/mfg01-did-sewing-and-finishing-teams-meet-their-recorded-productivity-targets-equally-often)
- [CSV](https://www.crosstabs.com/examples/mfg01-did-sewing-and-finishing-teams-meet-their-recorded-productivity-targets-equally-often/download/csv) · [JSON](https://www.crosstabs.com/examples/mfg01-did-sewing-and-finishing-teams-meet-their-recorded-productivity-targets-equally-often/download/json)
