Posts Tagged ‘polls’

h1

MONDAY MORNING QUARTERBACK DIGEST #33

November 5, 2018

Monday Morning Quarterback Marketing Digest. Tomorrow is a special day, we all get to vote! So, I want to suspend my usual blog and encourage everyone to vote! And don’t let polls influence you. You simply can’t trust polls, but you can trust voters. Here’s a few things you should know about polls:

• Polls are a snapshot in time – the way people feel at that particular moment. Things can change quickly – in fact, they can change within minutes before voting.

• Randomness is the key to polling and it is not easy. Are you getting a good sample? Today, that is very tough. A lot of younger people are rarely at home. Few people have land lines anymore. Plus, people are wary of callers who may be trying to sell them something. Our experience shows that only 10-20% of people will now respond to phone surveys. That is why we and others are using more online panels (like our Survey for Good).

• Respondents tend to think of a poll as a quiz. So, they want to give the right answer. They want to support social norms. For example, back in 2012, Gallup asks people if they would vote for a well-qualified candidate for President who happens to be a woman, 95% said yes. But would they? What the poll actually revealed was that people know what the social norms are: don’t endorse prejudice. In the same poll, 68% said they would vote for a gay or lesbian.

Yes, polling can be tricky. That’s why people ask for our help all the time.

One last thought, newspaper media endorsement (newspapers, TV, radio) usually don’t change people’s vote. According to research, they reinforce the opinions you have for a candidate or measure.

Have a great week and vote! Then, on Wednesday, be kind and gracious to others – no matter who won. If you voted, you showed up and that makes you special.

h1

MONDAY MORNING QUARTERBACK DIGEST #22

August 13, 2018

Monday Morning Quarterback Marketing Digest. Here are some marketing and research thoughts for this week. Email me (Mark@dennettgroup.com) if you need more info on any subject.

DO YOUR CUSTOMERS HAVE MORE MONEY TO SPEND? – Not really. Yes, unemployment is as low as it’s been in nearly two decades. But research reveals that today’s real average wage (that is, the wage after accounting for inflation) has about the same purchasing power it did 40 years ago. Wage gains have mostly been to the highest-paid workers. So, your customers really do not have more money to spend on your products and services.

BUILDING A BETTER VISITOR WEBSITE – Are you considering updating or revising your visitor website? If so, you will want a copy of our new white paper “KEYS TO MAKING WEB VISITORS, REAL VISITORS.” This report shares findings from various website content studies including our six-year study of Oregon visitor websites. While much of the content of our study is confidential (only shared with sponsors) the new white paper reveals what is most important to website visitors in influencing them to become real visitors. Cost of this study is $95. Contact us if you are interested. Our next website study launches in October and we have room for more partners. Join us and the white paper is free.

ARE ONLINE POLLS ACCURATE? – The probability-based phone poll has been the gold standard for public opinion surveys. However, more polls are now using online nonprobability samples, online “opt-in” panels. The “Survey for Good” Jackson County panel that we are recruiting for (www.DCGResearch.com) is an example of an opt-in panel. How accurate are these polls? According to a 2016 PEW Research Center Study, based on over 30,000 interviews, they can be accurate if you apply statistical sampling techniques.

Because online panel members self-select, your samples will not resemble the larger population. Therefore, you need to employ statistical techniques, such as raking, propensity weighting and matching, to make sure your sample is projectable. Bottom line: If you are considering using an online panel, which are available from many vendors (such as Survey Monkey), it is critical to have outside research help in developing questions and applying selected statistical techniques. Contact me if you would like to learn more about our research planning and review services.

Until next week, enjoy summer and have a great day.

h1

MONDAY MORNING QUARTERBACK DIGEST #21

August 6, 2018

Monday Morning Quarterback Marketing Digest. Here are some marketing and research thoughts for this week. Email me (Mark@dennettgroup.com) if you need more info on any subject.

ARE YOU REDESIGNING YOUR VISITOR WEBSITE? – As summer tourism season winds down, many Destination Marketing Organizations (DMOs) are looking at updating their websites. You might want to discover the website content insights we have learned from our six-year study of DMO websites. The study reveals what is most important to website visitors in influencing them to become real visitors. Contact us if you are interested in a brief analysis of your website, prior to design work. Our next DMO website study launches in October and we have room for a couple more partners.

WHEN IS A MAJORITY, NOT A MAJORITY? – One of my pet peeves is how people use the word majority. Polls (quantitative research) are based on sampling, not on a survey of the entire (true) population. This reality can create challenges when sharing results. To better appreciate the problem, the Pew Research Center looked at three recent news headlines:

Majority of Canadians would boycott U.S. in response to trade war
A majority of travelers are picking destinations based on food
Majority of Americans approve of Trump’s handling of the economy for the first time

Each headline uses majority, but it does not mean the same thing. In the first one, a “majority of Canadians” refers to 72%. In the second one, “a majority of travelers” refers to 64%. And in the third, a “majority of Americans” refers to 51%. Substantial differences!

In our research reports we never label a survey finding a “majority” unless our margin of error supports it. The margin of error describes the estimated range within which we would expect the answer to fall, if we surveyed everyone (the true population). For example, if a survey has +/- 3% margin of error with a 95% confidence level, and a 51% “majority” is reported, the true population value could fall between 48% and 54%. So, it is not a majority. But if 54% say “yes” – the true population value ranges from 51% to 57%. So, using the majority word is OK.

Question wording can also introduce error or bias. Another potential error is non response bias – i.e. people who respond are different in important ways from the people who don’t respond.

Bottom-line: Even if you are doing your own research, let us review your questions and methodology before you launch your study. Also, we can prepare or review your report and news release, so you do not get into public relations hot water.

MARKETING 101 – THIS DOG WON’T HUNT – Following up my post last week, it looks like MoviePass is finding common sense marketing. They will soon begin limiting customers (who pay $10 a month) to three movies a month. This is a change from its current allowance of one a day. The move will reduce its cash burn rate by more than 60% after it takes effect Aug. 15. They must be listening to an outside marketing consultant, finally.

Until next week, enjoy summer and have a great day.