This article is about the review and its’ importance. We
generally ignore the test cases review because of time constraint. As we know
testers duty is to identify the defects in the software developed. What happens
if they misunderstood the requirements? They will raise the wrong defects which
result in loss of quality development time. A review helps a lot to avoid this
tricky situation. So we should manage our schedule to have the review as a key
component, which plays a key role of the project success.The review is not only
limited to the test cases but from the requirements to test cases i.e.,
Business requirements understanding and analysis, Test scenarios, Traceability,
Completeness and correctness of the test cases.
Review is a static testing technique i.e., testing by not
executing the code. This can be done efficiently with the combination of human
and tools effort.
Ex: We can use the Microsoft Excel/Word for spell check and
grammatical errors. We can manually review the scenarios covered and test
design techniques like BVA and or ECP are used.
The below mentioned items should be reviewed in the same
priority order.
1.
Business Requirements
2.
Test Scenarios
3.
Test Cases
Business Requirements
Review:
Testing team should understand the business requirements
first in order to proceed with their regular testing activities. To have proper
understanding of the requirements, the requirements should be properly document
with expected level of details.
The requirements checklist assist in determining whether the
requirements are documented, traceable, correct, complete, unambiguous,
consistent, verifiable and approved.
Here I am concentrating from the testing point of view so
given only a subset of the overall checklist i.e., specific to the testing
team.
Entry Criteria: BRD/SID should be signed
off or near to sign off
First fill the below table then continue with the checklist.
Version
|
BRD/SID NAME
|
Author(s)
|
Reviewer(s)
|
Designation
|
Business requirements review Checklist.
ID
|
Items to Verify
|
Y/N/NA
|
Comments
|
1
|
Is each requirement uniquely and correctly
identified?
|
||
2
|
Is each requirement traceable to its source (including
derived requirements)?
|
||
3
|
Are all requirements written at a consistent and appropriate
level of detail?
|
||
4
|
Are individual requirements rated (or ranked), with
descriptions of priority provided?
|
||
5
|
Do the requirements provide an adequate basis for design
and system test?
|
||
6
|
Is each requirement verifiable by testing, demonstration,
review, or analysis?
|
||
7
|
Are all internal cross-references to other requirements
correct?
|
||
8
|
Does each functional requirement specify input and output,
as well as function, as appropriate?
|
||
9
|
Are validity checks on the
inputs defined?
|
||
10
|
Is the exact sequence of
operations described?
|
||
11
|
Have all dependencies on other systems been identified?
(Applications or application interfaces, databases, communications
subsystems, networking, etc.)
|
||
12
|
Are specific responses to
abnormal situations needed?
(e.g., overflow, communication
facilities, error handling/recovery)
|
||
13
|
Are any design or implementation constraints described?
|
||
14
|
Have business scenarios been constructed to illustrate (or
draw out) the requirements?
|
||
15
|
Are all reliability, recoverability (business continuity),
and performance requirements properly specified?
|
||
16
|
Are all security requirements properly specified?
|
||
17
|
Have all data privacy requirements been included?
|
||
18
|
Are time critical functions identified, and timing criteria specified
for them?
|
||
19
|
Have any regulatory, legislative, or standards-driven
requirements been addressed?
|
||
20
|
Have all quality attributes (characteristics) been
properly specified (i.e. efficiency, flexibility, interoperability,
maintainability, portability, reusability, usability, availability)
|
||
21
|
Are the requirements free of duplication and conflict with
other requirements?
|
||
22
|
Does each requirement have only one interpretation? If a
term could have multiple meanings, is it defined?
|
||
23
|
Are there measurable acceptance criteria for each
functional and non-functional requirement?
|
||
24
|
Is each requirement in scope for the project?
|
||
25
|
Are all requirements actually requirements,
not design or implementation solutions?
|
||
26
|
Are the time-critical functions identified,
and timing criteria specified for them?
|
||
27
|
Have internationalization issues been
adequately addressed?
|
Exit Criteria:
Repeat the review of business requirements till most of the
checklist responses say ‘Y’ and remaining ‘NA’ with supporting comments.
Test Scenarios Review:
Once
the business requirements understanding and review is completed tester by coordinating
with the lead should prepare the test strategy for that BRD/SID. Based on the
strategy tester should go for test scenarios preparation. Once the scenarios
are completed test lead/BA/SA or appropriate person should take care of the
scenarios review with below mentioned check list.
The
scenarios checklist assists in determining whether all the business requirements
are covered in the scenarios and are documented, traceable, correct, complete,
unambiguous, consistent, verifiable and approved.
Entry Criteria:
BRD/SID should be reviewed
Test Strategy should be reviewed
First fill the below table then continue with the checklist.
Version
|
BRD/SID/Strategy/Scenario Document
Name
|
Author(s)
|
Reviewer(s)
|
Designation
|
Scenarios review Checklist.
ID
|
Items to Verify
|
Y/N/NA
|
Comments
|
1
|
Is the document spell checked?
|
||
2
|
Is the correct version of the BRD/SID referred?
|
||
3
|
Has the correct template been used?
|
||
4
|
Does each scenario possess a unique identifier?
|
||
5 |
Is there at least one positive scenario exist for each
requirement?
|
||
6 |
Is there at least one negative (error handling/recovery or
exception) scenario exist for each requirement?
|
||
7 |
Is BVA technique used to write the scenarios?
|
||
8 |
Is ECP technique used to write scenarios?
|
||
9 |
Is the business requirement split in to maximum possible
scenarios?
|
||
10 |
Are all requirements covered and traceability exist?
|
||
11 |
Are all scenarios written at a consistent and appropriate level
of detail?
|
||
12
|
Is test data requirements identified?
|
||
13
|
Are the scenarios free of duplication?
|
||
14
|
Have the related areas that could possibly be affected by the
implementation of the requirement been identified and included in the test
cases
|
Exit Criteria:
Repeat the review of scenarios till most of the checklist
responses say ‘Y’ and remaining ‘NA’ with supporting comments.
Test Cases Review:
Once
the scenarios review is successful, tester prepares the test cases making sure
all the scenarios are covered and by making all the possible scenarios are
covered in one test case. Once the test cases are completed test lead or
appropriate person should take care of the test cases review with below
mentioned check list.
The
Test Cases checklist assists in determining whether all the scenarios are covered
in the test cases and are documented, traceable, correct, complete,
unambiguous, consistent, verifiable and approved.
Entry Criteria:
BRD/SID should be reviewed
Test Strategy should be reviewed
Scenarios should be reviewed
First fill the below table then continue with the checklist.
Version
|
BRD/SID/Strategy/Scenario/Test Case
Document Name
|
Author(s)
|
Reviewer(s)
|
Designation
|
ID
|
Items to Verify
|
Y/N/NA
|
Comments
|
1
|
Is the document spell checked?
|
||
2
|
Does each Test Case possess a unique identifier?
|
||
3
|
Are the Test Cases traceable backwards to requirements and
design?
|
||
4
|
Are the Test Cases traceable forward to test results?
|
||
5
|
Do the Test Cases specify the desired test procedures (steps,
actions, activities) for executing the tests?
|
||
6
|
Do the Test Cases define the input values required for test
execution?
|
||
7
|
Do the Test Cases define output values (expected results)
required for test execution?
|
||
8
|
Do the Test Cases record the actual results for the tests?
|
||
9
|
Do the Test Cases specify test data requirements?
|
||
10
|
Do the Test Cases specify pre-conditions for test execution?
|
||
11
|
Do the Test Cases specify post-conditions for test execution?
|
||
12
|
Do the Test Cases provide adequate space for capturing detailed
comments?
|
||
13
|
Do the Test Cases indicate whether the test passed or failed?
|
||
14
|
Has the Test data been embedded into the test cases?
|
Exit Criteria:
Repeat the review of test cases till most of the checklist
responses say ‘Y’ and remaining ‘NA’ with supporting comments.
A Review Findings Summary is a tool
created to document and track anomalies and issues identified during reviews.
The Review Findings Summary contains
the following information:
Definition
|
|
Review Type
|
Peer Review or Formal
Review
|
Artifact
|
The category of the
artifact under review, such as: Business Requirements Document, SID, Test
Scenarios and Test Cases
|
Author
|
The person who created
the work product under review.
|
Project
|
The official project
name.
|
Version
|
The version number of the
work product under review.
|
Date Review Started
|
The date of the review
meeting.
|
Date Review Closed
|
The date all anomalies,
issues and action items are closed.
|
Identifier
|
A unique identifier that
permits identification and sorting; suggested Project acronym + sequential
number (i.e., PROJ0001)
|
Anomaly Category
|
DC=Documentation Content,
TR=Traceability
|
Anomaly or Issue
|
Items identified and
described during the review.
|
Resolution
|
The solution for the
identified anomaly.
|
Date Resolved
|
The date an issue was
resolved and the Review Team agrees it was resolved correctly.
|
Status
|
The various states
through which an anomaly passes on the way to resolution and closure. The
anomaly states are:
·
Submitted – when an item is logged and reported for repair.
·
Assigned – when an item is assigned for repair.
·
Opened – when an anomaly is assigned for correction.
·
Deleted – when an item is originally reported as an anomaly, but
later deleted because the item is either a duplicate or not an anomaly.
·
Resolved - when an anomaly is corrected and sent for review or
verification.
·
Re-Opened – when an anomaly is closed and then reopened for
modification.
·
Returned - when an anomaly is reviewed, verified as
"incorrect", and returned to author.
·
Verified - when an anomaly is reviewed and verified as
"correct".
·
Closed - when an anomaly is successfully reviewed and closed
with a resolution and resolution date.
·
Deferred - when an anomaly is designated for correction at a
later date.
·
Duplicated – when an item is assessed to be a duplicate of a
prior record.
·
Escalated – when an item requires evaluation by management.
Note: The statuses listed
can be changed based on convenience.
|
Impact
|
They are classified in
three levels:
·
High Impact
·
Medium Impact
·
Low Impact
|
<Review Type>
Review Findings Summary
Artifact: Author: Project:
Version: Date Review Begun: Date Review Closed:
|
Anomaly Category
|
Anomaly or Issue
|
Date Resolved
|
Status
|
Impact
|
Anomaly or Issue:
Location:
Resolution:
|
|||||
Anomaly or Issue:
Location:
Resolution:
|
|||||
Anomaly or Issue:
Location:
Resolution:
|
|||||
Anomaly or Issue:
Location:
Resolution:
|
|||||
Anomaly or Issue:
Location:
Resolution:
|
Common Review Defects,
their Causes and suggested Preventive Measures:
The most common defects reported during the review of test cases
and the causes for each of them along with the suggested preventive actions are
tabulated below. Depending on what the review defects trend is in the project,
these actions can be implemented accordingly.
Defects
|
Causes
|
Preventive Actions
|
Incomplete Test Cases
Missed negative test
cases
|
·
Inadequate Functionality
Knowledge
·
Inadequate Testing
experience
·
Too Much or too little
information in the requirements
·
Oversight
·
Changes to requirements
after preparation of test cases
|
·
Provide application
specific Training
·
Provide training in
Testing concepts and methods
·
Do a thorough
Requirements review before test case preparation
·
Use Test Case Review
Checklist for all reviews
·
Periodic Requirements
review before submitting test cases for review
|
No Test Data
Inappropriate/Incorrect
Test data
|
·
Inadequate information in
the Requirements
·
Inadequate Functionality
Knowledge
|
·
Thorough Requirements
review before test case preparation
·
Provide application
specific Training
|
Incorrect Expected
behavior
|
·
Inadequate Functionality
Knowledge
·
Changes to requirements
after preparation of test cases
|
·
Provide application
specific Training
·
Do periodic Requirements
review before submitting test cases for review
|
Documentation Errors
Grammatical errors
Typos
Inconsistent tense/voice
Incomplete results/number
of test runs
Defect details not
updated
|
·
Oversight
·
Not attaching importance
to grammar checks
·
Not updating the test
cases with the actual results and defect details after every run.
|
·
Do a thorough spell check
before submitting the document for review
·
Provide training in
written communication
·
Discuss and reiterate the
process of updating the test cases for every test run
|
Changes to requirements
not updated in Test case
|
·
Changes to requirements
not communicated by business.
·
Not checking the comments
in the requirements document periodically
|
·
Request the Business
Analysts to pass on the information about changes to requirements.
·
Do periodic Requirements
review and update the test cases, when required before submitting test cases
for review
|
Possessing the knowledge of what could go wrong and how it could
be prevented will enable the team to prepare better test cases. This, coupled
with the usage of the review checklists in each phase would definitely pave way
to achieve effective and efficient review of test cases.
No comments:
Post a Comment