Reporting evaluation findings

Name:

Date:

Class:

Quick Questions

Shade in the A or B circle for each question.

1. What is the main purpose of an evaluation report?

A State how well each criterion was met, with evidence, and recommend improvements

B List every line of code that was written in the solution

2. Which conclusion for one criterion is best?

A Criterion met: the search ran in 0.8 s on 1000 records, under the 2 s target.

B The program seemed pretty fast and I think it was okay.

3. How should conclusions be organised in an evaluation report?

A One conclusion for each evaluation criterion

B One general conclusion that ignores the criteria

4. A criterion was only partly met. What should the report do?

A Leave it out so the report looks more successful

B Say it was partly met, give the evidence and recommend an improvement

5. Which is a good improvement recommendation?

A Make the program better.

B Add a range check to the age input, because 3 of 5 testers entered invalid values.

Match the Pairs

Draw a line connecting each item on the left to its match on the right.

Left column		Right column
1Criterion: Search returns results in under 2 seconds. Result: Met. Evidence: 9 of 10 test searches took under 1.5 seconds.		AThe evidence is an opinion, not measured data or user feedback. The conclusion is not supported.
2Recommendation: Add an index to the student table so that searches stay fast when the data grows.		BCriterion 3 was Partly met, so: Recommend adding error messages that tell users how to fix invalid input.
3Each conclusion in an evaluation report must be supported by evidence, such as data from testing or user feedback.		CAn improvement: a specific change to the solution that is suggested because of the findings in the evaluation.
4An evaluation report links each recommended improvement to a criterion that was only partly met or not met.		DResult: Not met. Evidence: 6 of 10 users needed help to find the Save button in the survey.
5Criterion: The program is easy to use. Result: Met. Evidence: I think it is good.		EA conclusion for one criterion: it states whether the criterion was met and backs the conclusion with measured evidence.
6Result: Not met. Recommendation: Make the program better.		FThe recommendation is vague. It does not say what to change or link to the criterion that failed, and there is no evidence.

Definitions

evaluation report

Fill in the Blanks

Complete each line of code by filling in the missing parts.

target = 2.0 actual = 1.4 met = actual __ target

finding = {"criterion": "Load time", "________": "Avg 1.4 s in 10 tests", "conclusion": "Met"}

finding = {"criterion": "Load time", "evidence": "Avg 1.4 s in 10 tests", "__________": "Met"}

finding = {"criterion": "Error messages", "conclusion": "Partly met", "___________": "Add fix tips"}

if ___ met: print("Improvement: speed up search")

Spot the Error

Shade in the letter of the line that contains an error.

Question 1

A Criterion: Login takes < 5 s

B Evidence: avg 4.2 s, 8 tests

C Conclusion: Met

D Improvement: Login was fast

Question 2

A Criterion: Error rate < 2%

B Evidence: 6% of entries failed

C Conclusion: Met

D Improvement: Add range checks

Question 3

A Criterion: Easy to navigate

B Evidence: I feel it is easy

C Conclusion: Met

D Improvement: Add a search box

Question 4

A Criterion: Report runs < 10 s

B Evidence: Colours look nice

C Conclusion: Met

D Improvement: Index the table

Question 5

A Criterion: Search is good

B Evidence: avg 1.4 s, 10 runs

C Conclusion: Met

Answer Key

Reporting evaluation findings

Section A: Quick Questions

A — State how well each criterion was met, with evidence, and recommend improvements

A — Criterion met: the search ran in 0.8 s on 1000 records, under the 2 s target.

A — One conclusion for each evaluation criterion

B — Say it was partly met, give the evidence and recommend an improvement

B — Add a range check to the age input, because 3 of 5 testers entered invalid values.

Section B: Match the Pairs

Criterion: Search returns results in under 2 seconds. Result: Met. Evidence: 9 of 10 test searches took under 1.5 seconds. → A conclusion for one criterion: it states whether the criterion was met and backs the conclusion with measured evidence.

Recommendation: Add an index to the student table so that searches stay fast when the data grows. → An improvement: a specific change to the solution that is suggested because of the findings in the evaluation.

Each conclusion in an evaluation report must be supported by evidence, such as data from testing or user feedback. → Result: Not met. Evidence: 6 of 10 users needed help to find the Save button in the survey.

An evaluation report links each recommended improvement to a criterion that was only partly met or not met. → Criterion 3 was Partly met, so: Recommend adding error messages that tell users how to fix invalid input.

Criterion: The program is easy to use. Result: Met. Evidence: I think it is good. → The evidence is an opinion, not measured data or user feedback. The conclusion is not supported.

Result: Not met. Recommendation: Make the program better. → The recommendation is vague. It does not say what to change or link to the criterion that failed, and there is no evidence.

Section C: Definitions

evaluation report — A document that states how well each criterion was met, gives evidence for each conclusion, and recommends improvements to the solution.

Section D: Fill in the Blanks

target = 2.0
actual = 1.4
met = actual <= target

finding = {"criterion": "Load time",
    "evidence": "Avg 1.4 s in 10 tests",
    "conclusion": "Met"}

finding = {"criterion": "Load time",
    "evidence": "Avg 1.4 s in 10 tests",
    "conclusion": "Met"}

finding = {"criterion": "Error messages",
    "conclusion": "Partly met",
    "improvement": "Add fix tips"}

if not met:
    print("Improvement: speed up search")

Section E: Spot the Error

D — An improvement must recommend a change to the solution, such as caching user data. 'Login was fast' only repeats the result.

C — The evidence shows a 6% error rate, which is above the 2% limit. The conclusion should be 'Not met'.

B — Evidence must be data gathered from testing or users, such as 9/10 found the menu. A personal opinion is not evidence.

B — The evidence must relate to the criterion. This criterion is about speed, so the evidence should be timing results, not the colours.

A — A criterion must be measurable so a conclusion can be judged against it. 'Is good' is vague. It should state a target such as 'under 2 s'.