The art of asking survey questions final

The Art of Asking
A Publication of
SURVEY
QUESTIONS
&

4. Introduction: The Art of Asking the Right Questions.
7. Survey Yourself Before Others: 5 Questions to Ask Yourself
Before You Start Writing.
CONTENTS.
Keep an eye out for
hyperlinks to other
resources to learn
more, too.
Click the text
above to jump
to a section.
27. The Do’s of Writing a Sound Survey.
47. The Don'ts of Writing a Sound Survey.
69. Wrap Up: Getting Started With Your Own Survey.

AUTHORS.
Sheila Grady is the Marketing Programs Manager at SurveyMonkey. An expert in survey
design, she works with customers, partners, and media to develop original survey-based
content. Additionally, she creates educational content for SurveyMonkey customers in a variety
of mediums including videos, webinars, e-books, blog posts, and social media.
Shannon Johnson is a Content Strategist at HubSpot, where she occasionally works with partners
like SurveyMonkey to create marketing resources you might actually want to download, read, and
share. Shannon has written, designed, and published dozens of ebooks, guides, SlideShare
presentations, blog posts, and other resources on various topics related to inbound marketing, social
media, and content creation.
Co-authored
& designed by
@shannopop
Co-authored by
@sheilacgrady

The Art of Asking
THE RIGHT QUESTIONS

When you need to gather information from an entire population, the
best way to do so is to send a survey. Problem is, surveys are difficult
to write. There are so many ways to ask questions, and if you don’t ask the
right ones, you won’t get the information you need.
That’s why HubSpot and SurveyMonkey collaborated to create this
guide – to teach you the art of asking the right questions so that writing
your next survey is both painless and productive.
Let’s get started.

First things first: Before you survey others, you need to take an honest
survey yourself.
If you’re getting ready to send an online survey, you probably already
have some questions in mind for your audience. No doubt about it —
what you ask your respondents should be well thought out. Equally
important, though, are the questions you ask yourself before you send
a survey. Five basic questions — why, who, what, how, and when — don’t get
as much attention as the questions you include in your survey. But they
should. Take time to answer them. Your results will thank you for it.

5 Questions to Ask
BEFORE WRITING A
SURVEY

The most important step in a survey is figuring out what you actually want to
know. It’s important to make your objectives really clear up front or the
next steps of survey formation won’t go so well.
For example, stay away from vague goals like measuring “satisfaction.”
What do you mean by satisfaction? Do you want to know if your public
awareness campaign was engaging and fun? Or if it was
informational? Or persuasive? Focus on your end goal: What decision do you
need to make or behavior do you need to change based on the
feedback you get from your survey?

Not only will having a clear sense
of the feedback you want help
you to analyze your data when
you get it back; it will also make
writing your actual survey questions clearer
and simpler — something that the
people answering your survey will
definitely appreciate. Which
brings us to our next question …

This may seem like a silly question, but it really is very important.
Survey respondents should be a “sample” of a “population.”
• A population is the entire set of people you want to ask – American
citizens, non-profit workers in Texas, or teenage internet users in
New York City.
• Your sample is the portion of that bigger population that actually ends
up taking your survey.

You can think of a population as all the
fish swimming around in the Pacific
Ocean and a sample as all the fish you
caught on Tuesday afternoon. Figuring
out which fish you want to catch is
directly related to why you’re fishing.
That is, figuring out who you want to answer
your questions is directly related to why you’re
asking in the first place.

Let’s say you’re a non-profit looking to make sure you get lots of
donations for your next year.
Do you want to make sure your current donors are happy with the way
you are spending their contributions? Or are you trying to figure out
how to attract new donors? If it’s about your current donors, then your
population would be all of your current donors, and to access that
population, just send your survey out to your mailing list of donors.
Your sample will be the people who actually respond to your survey.

If you’re trying to get new donors, however, it gets a little trickier. First
you have to spend a little more time defining who your population is
exactly. That is, who could potentially donate to your organization?
Adults probably. Maybe only English-speaking adults. If you’re a local,
grassroots organization, perhaps only American adults who live in San
Francisco. It’s really up to you. If you need help accessing a sample of
your target population, SurveyMonkey has an Audience available.
Now that we’ve tackled the question of who to ask, let’s move on to
our next question …

You’ve got your objectives and your sample
population nailed down. But your survey is still
blank. What do you ask?!
Well, that’s a complicated problem, and there’s
not just one right answer. The short answer?
Ask questions that measure the things you care about. If you
have a clear sense of the feedback you need
to get in order to reach your end goal,
determining what to ask will be a lot easier.

What makes writing surveys so difficult is that you have to operationalize a
fuzzy concept, which essentially means you need to figure out how to ask
questions in a way that they can measure an intangible, fuzzy concept
like happiness, for example. Some questions you could ask to measure
how happy someone is include:
• How many times did you laugh today?
• How optimistic are you about your future?
• How stressed were you this week?

Neither question is right nor
wrong. The questions you choose
to ask should ultimately provide
you with the information you need
to measure your goal. If you can’t do
something with the information,
don’t measure it!

There are so many different kinds of question designs and factors to
consider when producing unbiased and relevant survey responses.
To help you understand the range of questions you could ask, take a
look at SurveyMonkey’s overview of the types of survey questions you
can add, and view a demo of how they’ll appear in an actual survey.
These range from open-ended questions (comments or essays) to
closed-ended (yes/no, multiple choice, rating scale, etc.), and also
have a variety of formatting options.

The most important thing to consider when writing your survey
questions is to keep your questions simple, straightforward, and concise. This
will make it easy for your survey-takers to understand what you’re
asking, and will make it easy for you to analyze your data.
If you’re not sure if your questions are easy enough to understand, test
them out on a friend or someone who’s not familiar with your particular
industry to make sure the questions are comprehensive.

To understand when you should conduct your survey, let’s revisit the
fishing metaphor. When you go out fishing, how many fish you catch
can depend on what time you go out to fish and how long you stay out
there. Some kinds of fish tend to be easier to catch in the morning,
others at night. Fishing for 2 hours will yield a different catch than
fishing for 2 days.
In surveys, depending on who you want to “catch,” you’re going to
want to send out invitations at different points of the week, and leave
surveys open for different amounts of time — especially if you want to

get more responses.
Closing a survey too quickly can frustrate people who tried to respond,
and exclude people who are just a little slower at getting around to
things. This could potentially bias the conclusions you draw in an
unhelpful way.
For example, if you’re trying to recruit new donors, you might be
interested in what would motivate people who don’t tend to be quick
responders.

So, there you have it. Use these 5
questions to guide you, and your survey
will be off to a good start before you’ve
even started writing it! And that’s what
we’re going to address next in this guide:
what are the basic do’s and don’ts for the
actual writing stage of conducting an
effective survey?
Let’s dive in.

The Do’s of Writing
A SOUND SURVEY

A good survey should flow in an
orderly fashion, help to stimulate recall
(if necessary); and motivate the
respondent to reply. That’s why it’s
good practice to follow these
survey “do’s,” and avoid
unintentional violation of the
survey's objectivity by steering
clear of the “don’ts” that follow.

BE Clear ABOUT YOUR INTENTIONS.1

Remember your “why?” – the entire reason you were writing a survey in
the first place? Communicate that to your audience so they can
identify with your purpose.
If your survey doesn’t seem to have a purpose, good luck trying to get
anyone to take it! At the beginning of your survey, start with a short
introduction like, “Thank you for taking the time to complete this survey.
We value your feedback so we can improve our service. Your
responses will remain confidential.”

Crafting short surveys helps ensure that your respondents start and finish
your survey. Respondents are much more receptive to partaking if they
know your survey won’t take them much time to complete.
Respondents take more time to respond to each question when taking
shorter surveys as well, which means you’re more likely to collect
accurate feedback. Brevity also forces you to focus on your “why” –
the reason you’re writing and distributing the survey in the first place. It
helps you think about the most critical questions you could possibly ask to
retrieve the answers you need.

USE Words INSTEAD OF NUMBERS.3

If you’re going to give your survey-
takers answer choices, try not to use
too many choices for any given
question. People get overwhelmed. For
rating questions, make sure to label the
answer choices and limit the number of
options to no more than seven—the
most number of choices people can
typically consider at once.

So that means don’t ask your survey-takers to rate how happy they are
on a scale of 1 to 5. Ask them if they’re extremely happy, very happy,
moderately happy, slightly happy, or not at all happy. Words are easier for
people to think about than numbers.
If you’re trying to measure someone’s attitudes or behaviors, consider
using a Likert scale question. It's one of the most popular (and reliable)
ways to measure opinion from one extreme to another (for example,
not at all likely to extremely likely).

USE OPEN-ENDED QUESTIONS
Sparingly.
4

Asking open-ended questions, such as “What do you like about your
boss?” can be great for qualitative research. The problem with
questions like these, however, is that they’re unfocused and therefore hard to
analyze since the responses you’ll get will be so varied. In addition to
being difficult to analyze, the questions themselves are easy for
respondents to skip since they take more time and thought to answer.
Use open-ended questions sparingly (we recommend no more than
one per survey) to save yourself time in the analysis stage.

Sometimes it can be tempting to craft questions that provide
respondents with a simple “yes” or “no” answer because they’re really
easy for respondents to answer, and it would seem as if the 50/50
responses would make for easier analysis later on.
Even though “yes” or “no” questions are appealing, try to use them
sparingly for two major reasons: People’s views and behaviors can be
fuzzy, so their opinions about what you’re measuring don’t always fall
into one of two buckets.

Secondly, writing questions that allow your
respondents’ opinions to be seen in various
shades of gray from “yes” and “no,” to
“agree” and “disagree” (e.g. Likert scale
questions) actually ensures there’s enough
variance for statistical analysis, and gives
you more information because it not only
tells you what they feel, but how strongly they feel
about something.

ONLY ASK QUESTIONS THAT GIVE
YOU ANSWERS YOU CAN Act On.
6

The whole point of conducting a survey is to find out a key piece of
information in order to actually change or influence some sort of action
or behavior. When you conduct a survey to retrieve answers that you
have no immediate plans to do anything with, you waste yours and
your respondents’ time and energy.
Ask questions that give you actionable data so you can do something with the
results, and be prepared to make some radical shifts based on the data
you receive – not just the data you hope to get.

Start Broad, THEN GET NARROW.7

A general, logical, overall flow is important, because the order that specific
questions appear in your survey can directly impact the responses you
gather. This is why it is best to start broad and get narrow with your
questions.
Start with the most general question to get a clean reading on your
survey topic. For example, in an event feedback survey, the
respondents’ overall impression of the event should come first. That
way you know their reaction is based on their general opinion of the
event and not biased by the specifics like the event catering or venue.

Follow Up WITH YOUR RESPONDENTS.8

A great survey is like a great conversation in that it’s written in a
personable, logical fashion; it is written in a tone the audience can
relate to; and it allows the respondent to feel listened to. Oftentimes,
this is the sole reason why respondents choose to participate in
surveys in the first place: They want to be heard. If someone took the time to
answer your questions, it’s a good idea to follow up on your
“conversation” – even if it’s a more general summary – indicating what
action was taken.
Now that you know what to do, let’s move on to what not to do.

The Don'ts of Writing
A SOUND SURVEY

DON’T WRITE Leading Questions.1

For example: We have recently upgraded SurveyMonkey to become a first
class tool. What are your thoughts on the first class site?
You don't want to lead your respondents into answering a certain way
based on the wording of the questions.
By indicating that SurveyMonkey is a first class tool, you’re
encouraging respondents to respond positively to the site upgrade.

What are your thoughts on the changes to SurveyMonkey?
Instead, replace with something like this:
Removing the bias allows respondents to answer more objectively.
They get to focus on what they like or don’t like about the site changes
without the survey designer steering them one way or another.

DON’T WRITE Loaded Questions.2

For example: How terrible is dairy for your health?
Loaded questions contain emotionally charged assumptions that can
push respondents towards a specific answer choice. By choosing that
answer choice, they thereby agree to the question’s presupposition.
By answering this question, the respondent must agree that dairy is
unhealthy.

Do you think dairy is healthy, unhealthy, or neither?
Instead, consider something like this:
The respondent gets to determine whether dairy is unhealthy or not by
choosing from a balanced set of options.

For example: What do you think of the latest feature added to the HubSpot’s
all-in-one inbound marketing platform?
Don’t build in assumptions into your questions. Avoid asking questions
that assume the respondents are familiar with the specifics of it.
Include details or additional information if necessary.
We’re assuming you not only know HubSpot sells inbound marketing
software, but that you’re also up to date on all its latest updates.

HubSpot recently launched Campaigns, an app that allows you to track all of
your inbound marketing efforts back to identified campaign goals. How likely
are you to try the Campaigns app?
Instead, try something like this:
This version of the question provides enough context as to what the
Campaigns app is before asking whether the respondent would be
likely to experiment with it.

For example: Who was central to your ontogeny?
Use simple words that are direct and familiar. Try not to use jargon or
technical concepts, as it may make your respondents feel unintelligent,
or have to pause to consult the dictionary.
Using “ontogeny” instead of a well-known synonym may confuse your
respondents, and confused respondents provide inaccurate data.

Who was central to your development?
Instead, consider something like this:
See what we mean? Much better. Why use “ontogeny” when you can
just use “development?”

DON’T USE Double Negatives.5

For example: Which of these pictures is not unattractive?
Double negatives (negating adjectives) makes respondents have to
think harder, which makes them irritated and impatient, which means
they’re more likely to provide inaccurate answers (or quit altogether!).
Your respondent will be thinking, “Wait … are they asking if these are
attractive, or unattractive … or? Let me re-read …”

Which of these pictures is attractive?
Instead, why not:
And voila! See how much easier this is to understand? The more direct,
the better.

DON’T WRITE Double-Barreled Questions.6

For example: How useful do you find SurveyMonkey's Help Center Topics
and blog?
Double-barreled questions ask two questions at once. In a survey
multiple ideas presented at the same time inhibits a useful response.
The respondent may find SurveyMonkey's Help Center Topics very
useful and the blog not as useful, but the question doesn’t provide
them any opportunity to distinguish their opinions on either.

Q1: How useful do you find SurveyMonkey's Help Center Topics?
Q2: How useful do you find SurveyMonkey's email support?
Instead, try:
Separating the questions into two allows the respondent to focus on
the helpfulness of each resource.

When you give respondents the opportunity to answer a question with
“no opinion,” “I don’t know,” or “not applicable,” you’re communicating
that it’s okay to not provide an opinion, which defeats the whole
purpose of you conducting the survey. Force respondents to provide an opinion
by doing the hard work up front and ensure you’ve provided them with
sufficient responses to choose from.
Eliminate questions that include “no opinion” as a response, only use “I
don’t know” as a possible response for really difficult questions, and
instead of using “not applicable” as a response option, use skip logic.

“How useful is our blog?” with “Not applicable” …
For example, instead of pairing:
Consider asking a “yes” or “no” question such as:
(Assessing past behavior is an example of when “yes” or “no”
questions make sense: Either they’ve visited the blog or they haven’t).
“Have you ever visited our blog?” and use skip logic to direct respondents to
applicable follow up questions.

Wrap Up:
GETTING STARTED
WITH YOUR OWN
SURVEY.

The main point we’re trying to get across is this: Don’t get lazy when designing
your survey. Do the hard work for your respondents so they can breeze
through your survey while providing you with quality data.
If your survey doesn’t seem to have a meaningful purpose, if it doesn’t
flow in a logical order, if it contains complicated language or tedious
question types (e.g. ranking or matrix questions), or if it’s for whatever
reason difficult to understand, you’ll annoy your respondents and you’ll
get data that you can’t do much with. And that defeats the whole
purpose of conducting a survey in the first place.

Remember: Bad data can be worse than no data at all. Bad data leads to bad
decisions, which can end up costing your company lots of money.
So, now that you’ve learned some of the best practices associated
with creating survey from scratch, take a stab at creating your own
survey using the workbook we provided.
Once complete, you should be well on your way to distributing a
survey that will help you tackle the problem at hand.
Now, what problem do you want to solve?

Want the right answers?
Create surveys.
Get feedback.
Get access to survey templates and other
tools that make survey design simple.
Sign Up for a Free Account

What is ?
All-in-one inbound marketing software.
Get a Demo Video Overview
Get a Free Marketing Assessment

The art of asking survey questions final

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to The art of asking survey questions final (20)

More from Safe Rise (20)

Recently uploaded (20)

The art of asking survey questions final