Unit 3: Effort Estimation - Practice Quiz

1 What does effort estimation determine in software project management?

2 Which unit is commonly used to express software development effort?

3 Why are estimates made at the beginning of a software project often uncertain?

4 Which situation can make historical estimates unreliable for a new project?

5 Which estimation technique relies mainly on the experience of knowledgeable professionals?

6 What is analogy-based estimation?

7 What does Albrecht Function Point Analysis primarily measure?

8 Which item is counted in Albrecht Function Point Analysis?

9 What is the main purpose of Functions Mark II?

10 Functions Mark II mainly analyzes software as a collection of what?

11 What does COCOMO stand for?

12 What primary size measure is used by the basic COCOMO model?

13 Which COCOMO extension includes cost drivers such as personnel and product attributes?

14 Which version of COCOMO was designed to better support modern software development practices?

15 What can an AI-based effort estimation tool learn from?

16 Which technology is commonly used in AI-based effort estimation?

17 What type of information does NLP-based requirement analysis primarily process?

18 Which requirement problem can an NLP tool help identify?

19 Which historical value is most useful as a target when training an automated effort estimation model?

20 Why should historical project data be cleaned before automated estimation?

21 A project manager estimates that a system will require 24 person-months of effort. What does this estimate primarily represent?

22 A project has an estimated effort of 36 person-months. Why might assigning 12 developers not reduce its duration to exactly 3 months?

23 A team estimates a new mobile banking system using productivity data from a small internal reporting tool. What is the main weakness in this estimation basis?

24 Requirements are incomplete, but management asks for a single precise effort estimate. Which response best handles the uncertainty?

25 A company must estimate a project that resembles three completed projects with reliable effort records. Which technique is most directly applicable?

26 Three experts independently estimate a task before discussing their assumptions and revising their values over several rounds. Which technique are they applying?

27 A system has 8 low-complexity external inputs and 5 average-complexity external outputs. Using weights 3 and 5 respectively, what is their combined unadjusted function point contribution?

28 A project has 200 unadjusted function points and a total general system characteristic score of 40. Using , what is the adjusted function point count?

29 Why can Function Point Analysis be used before the programming language is selected?

30 A Mark II transaction contains 10 input data element types, 6 entity types referenced, and 8 output data element types. Using weights , , and , what is its functional size?

31 A business transaction reads several entity types but produces only one small output. Under Functions Mark II, which change would most increase its measured size?

32 Using Basic COCOMO for an organic project, . Approximately how much effort is required for a 32 KLOC system?

33 A small experienced team is developing a relatively flexible business application in a familiar environment. Which Basic COCOMO development mode is most appropriate?

34 Basic COCOMO estimates an organic project's effort as 100 person-months. Using , what is the approximate development duration?

35 A nominal COCOMO estimate is 80 person-months, and the combined effort adjustment factor is 1.25. What is the adjusted effort?

36 Two COCOMO II projects have the same size and effort multipliers, but Project X has scale factors that produce a larger exponent. What is the expected effect?

37 An AI estimation tool performs well on training projects but poorly on new projects. Which action most directly checks whether the model is overfitting?

38 A neural estimation model predicts 500 person-hours, but managers cannot determine which project attributes drove the result. Which concern is most relevant?

39 An NLP tool flags the requirement "The system should respond quickly" during automated analysis. What issue has it most likely detected?

40 A company trains an automated estimator using historical project records from different teams. Before training, what is the most useful data preparation step?

41 A project is estimated to require 48 person-months of effort. Management assigns 12 developers and concludes that delivery will take exactly 4 months. Which interpretation is most defensible?

42 A company estimates a cloud-native project using productivity from older mainframe projects. The historical projects report effort per delivered source line, while the new project relies heavily on managed services and generated code. What is the most serious basis problem?

43 An estimation repository contains only projects that reached production; canceled and severely delayed projects were removed during data cleaning. A model trained on this repository consistently underestimates high-risk projects. Which issue best explains the behavior?

44 A team wants to use Wideband Delphi, but the technical lead publicly announces a 900-hour estimate before participants make their own estimates. Which modification most directly reduces the resulting anchoring risk?

45 Two independent sequential activities use PERT estimates. Activity A has weeks and activity B has weeks. Using the PERT mean and variance and a normal approximation, what is the approximate one-sided 95th-percentile completion time?

46 An application has 10 low-complexity EIs, 6 average EOs, 4 high EQs, 3 average ILFs, and 2 low EIFs. Using standard IFPUG weights and a total general-system-characteristic score of 40, what is the adjusted function point count?

47 A target application reads customer records from a CRM database maintained entirely outside the target application's boundary. The target application neither creates nor updates those records. How should the customer data be classified for the target application?

48 A system initially has 200 unadjusted function points and a value adjustment factor of . A new high-complexity external output is added, while the general-system-characteristic score remains unchanged. What is the new adjusted size?

49 A Functions Mark II measurement uses , where is input data element types, is output data element types, and is entity-type references. For totals of 40, 25, and 70 respectively, what is the functional size?

50 Two systems expose the same number of user transactions, but one transaction references many entity types and returns substantially more output data elements. Why can Functions Mark II assign different sizes to the systems?

51 In Basic COCOMO's organic mode, effort is modeled as person-months, where is KLOC. If size doubles and all other assumptions remain unchanged, by approximately what factor does estimated effort increase?

52 Two projects use the same Intermediate COCOMO mode with exponent . Project X has 50 KLOC and an effort adjustment factor of ; project Y has 60 KLOC and a factor of . What is the approximate ratio ?

53 In a COCOMO II model, effort is , with . For a 100-KSLOC project, process improvements reduce by 5 while all effort multipliers remain unchanged. What is the approximate new-to-old effort ratio?

54 In COCOMO II, why is treating a scale factor as if it were an ordinary effort multiplier generally incorrect?

55 An AI effort estimator includes the final number of change requests as a training feature, even though that value is known only after project completion. Cross-validation shows excellent accuracy, but deployment performance is poor. What is the primary defect?

56 A neural estimator trained on small internal business applications is applied to safety-critical embedded projects. Its point predictions remain precise-looking even for inputs far outside the training distribution. Which control best addresses this risk?

57 Consider the requirement: "When the employee submits the form to the manager, the system shall archive it and notify them." Which NLP capability is most directly needed to determine what "it" and "them" refer to?

58 An NLP model predicts requirement complexity from individual requirement sentences. Sentences are randomly divided between training and test sets, so sentences from the same project occur in both sets. Why can the reported test accuracy be misleading?

59 An analogy estimator selects a 100-person-month historical project for a new project that is 20% larger in functional size. Historical analysis indicates effort scales with size exponent , and the new project's combined cost-driver multiplier is relative to the analogue. What effort should be predicted?

60 A historical effort model will be used each quarter while tools, staffing practices, and project types evolve over time. Which evaluation design best estimates future operational performance while exposing concept drift?