Piloting a survey before use

Name:

Date:

Class:

Quick Questions

Shade in the A or B circle for each question.

1. What is the main purpose of a pilot test of a survey?

A Try the survey on a few people first to find problems before the real collection

B Collect all the final data from the whole population

2. When should you pilot a survey?

A Before you send it to your real sample

B After all the real responses are collected

3. A pilot tester says, 'I didn't understand what 'regularly' means in Question 3.' What problem did the pilot test find?

A A missing answer option

B Unclear wording

4. Question: 'How do you get to school?' The options are Bus and Walk. A pilot tester cycles to school. What should you do?

A Add the missing options such as Bike and Car, plus Other

B Leave it, because most people will pick Bus or Walk

5. Your pilot testers take 25 minutes to finish a survey that was meant to take 5 minutes. What is the best action?

A Send it out anyway, because people will get used to it

B Shorten the survey before the real collection

Match the Pairs

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

Left column		Right column
1pilot test		AThere was no pilot test, so confusing questions and errors will only be found after the real data has already been collected badly.
2pilot_time = 6 (minutes)		BRecording how long a pilot tester takes to finish the survey, so you can check it is not too long
3A pilot tester says a question is confusing, so you rewrite the unclear wording before the real survey		CTrying a survey on a few people first to find confusing questions and errors before the real data collection begins
4Pilot testers find that no answer fits them, so you add the missing answer options		DQ3 old: How often do you use tech? Q3 new: How many hours a day do you use your phone?
5Plan: Send the survey to all 500 students without trying it on anyone first		EQ5 options: Bus, Train, Walk, Car, Other
6Pilot group: the 3 people who wrote the survey, so no changes needed		FThe pilot testers helped write the survey, so they already know what each question means and will not spot confusing wording or missing options.

Definitions

pilot test

Fill in the Blanks

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

=_______(B2:B6)

=_______(C2:C6,"Other")

=___(B2:B6)

=IF(B2_10,"Too long","OK")

=COUNTIF(_____,"Unclear")

Spot the Error

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

Question 1

A pilot_group ← 5 people

B give survey to pilot_group

C record minutes taken

D ignore confusing questions

E run real survey

Question 2

A options ← [Never, Sometimes]

B pilot_user picks an option

C IF none fits THEN ignore it

D ELSE keep options

Question 3

A start ← time at question 1

B end ← time at last question

C minutes ← end + start

D IF minutes > 10 THEN shorten

Question 4

A pilot_size ← 800

B real_size ← 200

C IF pilot_size < real_size

D OUTPUT 'Pilot is smaller'

E END IF

Question 5

A q ← 'Do you game much?'

B IF pilot says clear THEN

C q ← 'Hours gamed per week?'

D END IF

Answer Key

Piloting a survey before use

Section A: Quick Questions

A — Try the survey on a few people first to find problems before the real collection

A — Before you send it to your real sample

B — Unclear wording

A — Add the missing options such as Bike and Car, plus Other

B — Shorten the survey before the real collection

Section B: Match the Pairs

pilot test → Trying a survey on a few people first to find confusing questions and errors before the real data collection begins

pilot_time = 6 (minutes) → Recording how long a pilot tester takes to finish the survey, so you can check it is not too long

A pilot tester says a question is confusing, so you rewrite the unclear wording before the real survey → Q3 old: How often do you use tech? Q3 new: How many hours a day do you use your phone?

Pilot testers find that no answer fits them, so you add the missing answer options → Q5 options: Bus, Train, Walk, Car, Other

Plan: Send the survey to all 500 students without trying it on anyone first → There was no pilot test, so confusing questions and errors will only be found after the real data has already been collected badly.

Pilot group: the 3 people who wrote the survey, so no changes needed → The pilot testers helped write the survey, so they already know what each question means and will not spot confusing wording or missing options.

Section C: Definitions

pilot test — Trying a survey on a few people first to find confusing questions, missing options and timing problems before the real data collection begins.

Section D: Fill in the Blanks

=AVERAGE(B2:B6)

=COUNTIF(C2:C6,"Other")

=MAX(B2:B6)

=IF(B2>10,"Too long","OK")

=COUNTIF(D2:D6,"Unclear")

Section E: Spot the Error

D — A pilot test exists to find confusing questions so they can be fixed before the real survey. The line should be 'fix confusing questions', not ignore them.

C — If a pilot tester finds no answer option that fits, the options are incomplete. The fix is to add an option, not ignore the problem.

C — The time taken is the end time minus the start time. Adding them gives a meaningless number.

A — A pilot test uses only a few people. A pilot of 800 is bigger than the real sample of 200, so the IF check would fail. It should be a small number like 8.

B — The question should only be reworded when the pilot testers say it is unclear. The condition should be 'unclear', not 'clear'.