| Course | MT241 Sport Analytics |
|---|---|
| Unit | Unit 5 |
| Paper type | Data quality review |
| Length | About 1,110 words, 7 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | BS in Business Administration |
| Updated | October 2026 |
Free sample paper for MT241 Unit 5
Forty-One to Fifty-Two Percent Overnight: A Unit 5 Data Quality Review of a Club's Two-Season File
[Student Name]
Purdue University Global
MT241: Sport Analytics
Unit 5 Data Quality Review
[Instructor Name]
[Date]
The club, its staff, both seasons' figures and the vendor are composite and illustrative; the sources cited are real.
Why Review the File
Our composite club's analytics staff keeps one file for the 2024-25 and 2025-26 seasons, combining league play-by-play data, a vendor's expected-goals values and the club's own video tracking. The file feeds the weekly reports the coaches read. Before the staff uses it to compare the two seasons, this review asks of each field where it comes from, how it is recorded, whether that changed between seasons, and whether the field can be compared across them. In a survey of data users, Wang and Strong (1996) found that quality was rated on several dimensions besides accuracy, among them consistency, completeness and relevance to the task, and this review applies those three tests to each field.
The Field Table
The review covers six fields. Each entry states the source, the recording method, any change and a verdict.
Shot attempts: league scorers at each arena; recorded live; no change between seasons; usable for comparison.
Shot locations: league scorers; recorded live, with known differences between arenas; no change between seasons; usable, with care for home and road splits.
Expected goals: outside vendor; computed from league data; model version changed between seasons; not comparable until recomputed.
Hits: league scorers; recorded live; large differences between arenas; usable for road games only.
Giveaways: league scorers; recorded live; scorer judgment varies widely; not reliable for any comparison.
Zone entries: club's own video staff; coded from video after games; definition changed between seasons; not comparable until recoded.
Zone Entries: The Largest Change
Zone entries matter most because the coaches noticed the jump. In 2024-25 the club's video coordinator coded three kinds of entry: carried in, passed in, and dumped in. Only carries were counted as controlled. When a new coordinator took over in summer 2025, she merged carries and pass-ins into one controlled category, following the definition used in published tracking work, where a controlled entry is any entry with possession.
The review tested whether the change explains the jump. In the first season, pass-ins made up about 10 percent of entries. Adding them to the 41 percent of carries gives about 51 percent, almost exactly the second season's 52 percent. On the old definition, the second season's controlled entries would be about 42 percent, nearly unchanged.
The players did not change how they entered the zone; the spreadsheet changed what it called a carry.
The fix is simple. Because the second season's video is archived, the staff can recode 2025-26 under the old split, or recode 2024-25 under the new one. The review recommends the new definition, since it matches published work, and recoding the first season from the archived notes, which record pass-ins separately.
Hits: A Building Effect
The club recorded 26.4 hits per game at home and 19.1 on the road last season. The same players were on the ice, and the club's style did not change between home and road games. The difference comes from the scorers. Each arena's scorer decides what counts as a hit, and some count far more than others. Schuckers and Macdonald (2014) measured such differences across the league and found that rink effects for hits and similar events are large enough to distort team and player comparisons. The review recommends using road figures only, where the club faces many different scorers, or applying a published rink adjustment.
Expected Goals: Two Model Versions
The vendor's expected-goals values for 2024-25 were computed with an older version of its model. The vendor updated the model in summer 2025, adding inputs for pre-shot movement, and computed 2025-26 with the new version only. The two seasons' expected goals therefore measure slightly different things. The review asked the vendor to recompute the first season with the new model; until that file arrives, the expected-goals comparison between seasons should be marked as provisional.
Shot Locations: Usable With Care
Shot locations raise a smaller version of the hits problem. League scorers record where each shot was taken, and some arenas' scorers place shots slightly closer to the net or farther from it than others. The club's own arena records shots about two feet closer on average than the league's road arenas do, a small but steady difference. Because expected goals depend on distance, this shift slightly inflates the club's home expected goals for and against. The difference did not change between seasons, so comparisons from one season to the next are not affected. But home and road splits should be read with care, and any claim that the club creates better chances at home than on the road should be checked against road data first.
Giveaways and Missing Games
Giveaways are recorded by the same scorers as hits, with even more judgment involved, and the club's home figure is nearly twice its road figure. The review recommends not using giveaways for any comparison.
Three 2025-26 games are missing from the video file after a camera failure at the club's arena. The league play-by-play exists for those games, so shot attempts and locations are complete, but zone entries are not. The missing games are three of 82, so they are unlikely to change season averages, but any game-by-game chart should mark them as missing rather than leave a gap that looks like zero.
Recommendations
The review makes four recommendations. First, recode 2024-25 zone entries under the 2025-26 definition before comparing the two seasons. Second, use road games only for hits, or apply a rink adjustment, and drop giveaways from reports. Third, mark the expected-goals comparison as provisional until the vendor recomputes the first season. Fourth, keep a short record of every change in definition, staff or vendor model with the date it took effect, so the next review can trace changes quickly.
The review found nothing wrong with the club's shot attempt or shot location data, which come from the league and were recorded the same way in both seasons. Those fields can be used for comparison now.
What the Coaches Can Conclude
The coaches asked whether the players had changed how they enter the zone. On the evidence of this review, they have not, or not by much. Any claim of change should wait until the first season is recoded. Batini and Scannapieco (2016) describe data quality as fitness for a specific use, and for the coaches' use, comparing seasons, the zone entry field is not yet fit.
References
Batini, C., & Scannapieco, M. (2016). Data and information quality: Dimensions, principles and techniques. Springer. https://doi.org/10.1007/978-3-319-24106-7
Schuckers, M., & Macdonald, B. (2014). Accounting for rink effects in the National Hockey League's real time scoring system [Preprint]. arXiv. https://arxiv.org/abs/1412.1035
Wang, R. Y., & Strong, D. M. (1996). Beyond accuracy: What data quality means to data consumers. Journal of Management Information Systems, 12(4), 5-33. https://doi.org/10.1080/07421222.1996.11518099
The Purdue Global MT241 Unit 5 instructions
The fifth MT 241 unit asks students to review a data set before it is used for a decision. For each field, the prompt expects the source, how values are recorded, whether that changed over the period and whether the field can be used as intended. Test at least one suspicious change with numbers, showing whether a definition or recording change explains it. Data from scorers, vendors and in-house staff each carry different risks, so say who produced each field. Recommendations should be specific fixes, such as recoding, adjusting or dropping a field. Expect about five pages in APA 7, organized around a field table, with at least three sources on data quality or the specific measurement problems found.
How this MT241 Unit 5 example is built
The review opens on the two-season file and three tests drawn from Wang and Strong. The field table gives a one-line verdict for each of six fields, and a margin note explains why verdicts help a coach. The zone entry section tests the definition change with arithmetic, about 41 plus 10 equals the new 52, and the highlighted line places the change in the spreadsheet. Hits are read as an arena effect with Schuckers and Macdonald cited, and shot locations get a smaller version of the same caution, about two feet closer at home. The expected-goals section asks the vendor to recompute, marked in a note. Giveaways and the three missing games, four recommendations and a final answer to the coaches using Batini and Scannapieco close it.
Reading the MT241 Unit 5 grading rubric
A data quality review earns most of its marks for completeness, testing and practical recommendations. Each field should be traced to its source and recording method; a table that skips fields loses points. Testing earns more than suspicion, so a numerical check that a definition change explains a jump is worth more than noting that one occurred. Graders reward verdicts that say clearly whether a field can be used, and fixes that match each problem. Recognizing different kinds of error, definition changes, arena effects, model versions and missing data, shows range. A short record-keeping recommendation, correct citations and plain writing a coach could follow complete the rubric.
MT241 Unit 5 help: mistakes that cost points
Many reviews describe data in general terms without testing anything. Another frequent problem is treating every field as equally suspect, which gives the reader no guidance. Some students find an error and fix it in a way that creates a new one, such as adjusting home figures without a published method. Others forget missing records, or let a chart show missing games as zeros. Check whether the person who produced a field changed over the period. If you have a file but are unsure which fields are risky, our writers can help you build the field table and choose the test for each suspicious change.
Get an MT241 Unit 5 example written to your instructions
Two seasons, two leagues or two providers: whichever pair the Unit 5 review sets against each other, attach the file or its field list with the rubric. Within 24-48h, and free the first time, the review comes back field by field, each change tested where the data allows and a verdict on which comparisons still stand. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More MT241 papers and BS in Business Administration samples
- MT241 Unit 1: Discussion Board Post
- MT241 Unit 2: Metric Definition Exercise
- MT241 Unit 3: Descriptive Analysis
- MT241 Unit 4: Player Comparison Report
- MT241 Unit 6: Seminar Reflection
- MT241 Unit 7: Visualization Exercise
- MT241 Unit 8: Tactical Analysis
- MT241 Unit 9: Recruitment Recommendation
- MT241 Unit 10: Analytics Report
- MT240 Unit 5: Access and Participation Report
- MT237 Unit 5: Inventory Calculation
- MT236 Unit 5: Inventory Analysis
- MT221 Unit 5: Service Recovery Plan
MT241 Unit 5 questions, answered
What is scorer or arena bias in hockey data?
Differences in how official scorers in different buildings record the same kinds of events. Hits, giveaways, takeaways and even shot locations have shown arena-to-arena variation that the teams playing there cannot explain. Analysts often adjust for it or avoid comparing such counts across buildings, and a data quality review should check for it before trusting a home-road difference.
How can a review test whether a definition change moved a figure?
By recoding a sample of the earlier data under the new definition and measuring the difference. If recoding raises the rate in the sample by about as much as it rose between seasons, the rise is mostly definitional. The example recodes ten games, enough to estimate the effect without redoing a full season of video.
Is vendor-supplied data reliable enough for course assignments?
Usually, for the purpose it was built for, but not automatically across years or providers. Vendors update models, change collection methods and sometimes revise past values. A review should record which version produced each figure and avoid comparing numbers from different versions. Asking the vendor for recomputed history, or limiting comparisons to one season, are both defensible responses.
Where can I find a free MT241 Unit 5 sample paper?
The complete MT241 Unit 5 data quality review is on this page. It audits six fields in a two-season hockey file, tests a jump in controlled entries and recommends recoding, adjusting or dropping each problem field.
Why does the MT241 Unit 5 sample drop giveaways?
Giveaways depend heavily on each arena's scorer, and the club's home figure is nearly twice its road figure, so the review judges the field unreliable for any comparison.