Would I have your attention?

Thanks to my son, I recently became aware of an ingenious app called Logos Quiz which is currently the #5 ranked free iPhone app and the #6 ranked free iPad app. Like many games, this app is highly addictive.

The premise is simple: click on a partially-obscured logo and try to name the brand associated with it. If you name it, you get points. Repeat. The game contains over 500 brands, enough to keep anyone busy for some time.

Step 1

Step 2

Step 3

Step 4

This is gamification of market research in its purest form.

So which market research company created Logos Quiz?

Awkward silence.

The app was not created by a market research company. It was created by Javier Perez Estarriaga, a developer in Spain who, to my knowledge, has no prior connection to market research.

This is a classic example of disruptive innovation.

Market researchers should study Estarriaga’s example if they want to be on the right side of history.

And – just a thought – somebody might want to put a call in to this guy!

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

In 1997 Pew’s research yielded average response rates of 36%. By 2006, that figure was down to 21%. In 2012 it is down to 9%.

The 2012 figure was brought lower in part by the introduction of cell phone numbers to Pew’s samples; cell phone owners responded at a lower rate than landline owners.

It’s no secret that telephone response rates have been on a steady decline, so the Pew figures are not a complete surprise (I must admit, though, that there’s something jarring about seeing the figure dip into single digits.).

The question is, should you care?

The short answer is, “yes.”

As a person who cares about data, you should be concerned about the public’s willingness to participate in research.  People over time are putting less trust in societal institutions, including the survey research profession as a whole.

In this study Pew used multiple means to assess whether their survey results were impacted by lowered response; the conclusion was no, with proper weighting. They did find that survey participants tend to be more civic-minded than non-responders.

Ultimately, there is a limit to how low survey response can go before the ability to gain accurate readings on social phenomena are limited.

The long answer is, “it depends.”

Certainly, this news is more grave for people doing research in the public arena than it is for people in commercial market research, where online research is much more commonplace.

If most of your work is done using online panels or other sample sources, declining telephone response rates do not have as direct an impact on you.

However, declining survey participation is a phenomenon that extends well beyond telephone surveys. This is a societal trend that affects all research in one way or another.

In the online sphere, it is increasingly common for researchers to create their own “feedback communities” rather than relying on sample from traditional online access panels. This trend will surely continue, along with any other means there are to keep respondents engaged.

More research is needed to assess response rates in the online context; modeling a study after the Pew telephone investigation would be a great start.

Moreover, there should be more research on how the ability to survey respondents on their mobile devices affects the level and quality of survey response.

What do you think? Do you care about plummeting telephone response rates? Why or why not? Please share your opinions in the comments section below.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

Editor’s Note: Today’s post was originally published on the Survey Analytics blog.

Discrete choice conjoint analysis is a popular marketing research technique that helps you determine the optimal mix of features in a new product or service.

Survey Analytics invites you to a complimentary one-hour webinar, “How to Run Discrete Choice Conjoint Analysis.” The webinar is on Tuesday, May 22nd at 10am Pacific Time / 1pm Eastern Time / 6pm Greenwich Mean Time.

Click this link to register for the webinar.

The basics of discrete choice conjoint analysis are not hard to understand. In just one hour, you will gain an understanding of how to design, conduct and analyze a discrete choice conjoint analysis project.

This webinar will help you understand:

• What discrete choice conjoint analysis is
• The theory and logic behind discrete choice conjoint analysis
• When to use discrete choice conjoint in your research
• Specific case studies of how others have used discrete choice conjoint
• How to design a discrete choice conjoint project
• How to write a discrete choice conjoint questionnaire
• How to analyze the results of a discrete choice conjoint project

The webinar will feature Survey Analytics’ Enterprise Research Platform, with a newly enhanced discrete choice conjoint module, including such features as integrated d-optimal design generation, design import and upgraded part-worth calculations using maximum likelihood estimates.

The webinar will be conducted by Andrew Jeavons, President of Survey Analytics.

Anyone who attends the webinar or downloads the video and slides afterward will get one complimentary month of Survey Analytics’ discrete choice conjoint capability.

Click this link to register for the webinar.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

Congressman Daniel Webster (left), 19th Century Statesman Daniel Webster (right)

Legend has it there was a time in the history of the United States Congress when great statesmen debated the issues of the day.

19th Century Massachusetts Senator Daniel Webster is considered by historians to be one of the greatest statesmen in the history of the United States Congress.

Today we have another Daniel Webster in Congress. Congressman Daniel Webster currently represents the 8th District of Florida in the U.S. House of Representatives.

Unforunately, today’s Daniel Webster is no statesman. I could say harsher words, but that would be ungentlemanly.

Here is the definition of “statesman” (fittingly, from Webster’s Dictionary online).

Statesman: noun \ˈstāts-mən\
1. One versed in the principles or art of government; especially : one actively engaged in conducting the business of a government or in shaping its policies.
2. A wise, skillful, and respected political leader.

Technically, by the first definition, Congressman Webster fits the bill. However, he falls far short of the mark by the second definition.

What has Congressman Webster done do draw my ire?

He messed with my data.

Actually, he didn’t just mess with my data. He messed with your data. He messed with our data – data used by social scientists, economists and businesspeople to the benefit of our society.

But it goes even deeper than that. He messed with the very idea of the rightness of collecting and analyzing information for society’s benefit.

Specifically, Rep. Webster sponsored a successful amendment to a congressional funding bill to prohibit any funds from being spent on the U.S. Census Bureau’s American Community Survey. The ACS is the annual survey which collects a vast amount of information about American society; it is the legacy of the former decennial census “long form.”

The New York Times described Rep. Webster’s ploy as “know-nothingness at a new level.”

U.S. Census Director Robert Groves posted this erudite defense of federal statistics on his blog in a post entitled, “A Future Without Key Social and Economic Statistics for the Country.”

You see, there are two pernicious movements in American politics:

- one which casts aspersions on anything that has to do with knowledge, books, learning or education. These ideas are considered somehow the purview of elite bureaucrats and academics who care little about the common man.
- another which presupposes that the federal government is an Orwellian “big brother” trying to find out intrusive information about average people in order to bring about some sort of dystopian future.

I’m afraid Rep. Webster is trying to cash in politically on these movements. You see, he’s presumably an intelligent man.  Surely he understands the importance of collecting data on our society.

Yet he is willing to sacrifice the data which benefit America and the world for political expediency.

For the political pollsters out there, I say it is incumbent on you to defend the ACS and other federal statistical surveys to your clients in Congress and the federal government, particularly if they are like-minded with Rep. Webster.

Here are a few things regular people can do:

First, be aware. Here is a list of how each member of the U.S. House of Representatives voted on Rep. Webster’s amendment.

Second, make noise. Add your voices to mine, and speak out when you see powerful people disparaging data.

Third, take action. Americans, contact your government leaders to let them know you support the American Community Survey. Citizens of other countries, be on the lookout for similar political phenomena.

We should aspire to act like statesmen, even when our leaders do not.

Have a look at Rep. Webster’s statement as he introduced the amendment last week.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

Here are 5 of my favorite free online data tools provided by the U.S. Census Bureau.

1) The American FactFinder

http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml

The American Fact Finder is often the best place to start on the Census Bureau website. It is a user-friendly front-end to a vast amount of information made available by multiple Census Bureau initiatives, including the Decennial Census, the American Community Survey, and a number of economic surveys.

There is a helpful search interface that allows you to narrow down data by topic (people, housing, business, year, etc.), geography, race and ethnic group, and industry codes.

Or you can use the Quick Start search box to find data using just a few criteria – sort of like using the “I’m feeling lucky” button on Google.

You can also search your own (or any) street address to find out lots of relevant information, including county, town, block, census tract, metropolitan statistical area, congressional district, state legislative district, school district and more.

The American Fact Finder

2) The 2010 Census Interactive Population Map

http://2010.census.gov/2010census/popmap/

This is an interactive visualization of data from the 2010 decennial census.

You can interactively drill down on the map in order to see a map visualization of a particular community on a number of criteria, including total population, race, ethnicity, age, sex, and household characteristics.

The map shows various geographic levels, from states down to census tracts, color coded on criteria from dark (most populous) to light (least populous).

You can also choose up to 5 locations to compare them on the same criteria, and you can even embed the interactive data you select on a website.

2010 Census Interactive Population Map

3) County Business and Demographics Map

http://www.census.gov/cbdmap/

The County Business and Demographics Map is very similar to the Census Interactive Population map, except it depicts business data rather than population data.

The County Business and Demographics Map

4) DataFerrett

http://dataferrett.census.gov/

DataFerrett

In true federal style, the Ferrett in DataFerrett is an acronym (for Federated Electronic Research, Review and Extraction & Tabulation Tool). Coincidentially, it helps you ferret out the right data.

DataFerrett helps you identify, extract and recode data across federal, state and local surveys and statistical data, including both one-off and longitudinal datasets.

With DataFerrett you can search for the right variables and put them in something called a Data Basket.  From the Data Basket you can create spreadsheets, graphs, maps, PDFs and more right within the tool. And of course, you can download the data.

5) Censtats Databases

http://censtats.census.gov/

Looking to skip the bells and whistles and go straight for some of the most commonly requested data? The Censtats databases may be the place to look.

The Censtats databases are a good place to find quick tables on things like state and county population and business data. With the Censtats databases you can run these simple searches, and you can also download the relevant data files.

The Censtats databases also include information such as building permits, international trade data and housing data.

What about you? What are your favorite places to get data from the U.S. Census? How about from other agencies in the U.S. or other countries? What cool data tools can you point us to?

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

“When the individual’s behavior and consciousness get hooked to a routine sequence of external actions, he is a dead robot, and it is time for him to die and be reborn. Time to ‘drop out,’ ‘turn on,’ and ‘tune in.’

- Dr. Timothy Leary

OK, it’s no longer the 1960s, but the words of Dr. Timothy Leary are eerily prescient about the current state of survey research.

Allow me to explain.

When we expect respondents to take the same old boring surveys, when we misuse their time, and when we treat them with less than full respect, we cannot expect them to feel good about what they are doing.

However enthusiastic they may be about helping us help our clients, they eventually, understandably, become “dead robots,” responding in a rote fashion until they just can’t take it anymore.

Naturally, “dead robot” respondents provide lesser quality data, and, they are much more likely to drop out of our surveys.

We could all use tips and reminders about how to keep respondents “turned on” and “tuned in.” Here are 5 suggestions for you to consider using in your surveys.

And no, I’m not suggesting using any of Dr. Leary’s preferred methods.

Note: Ivana Taylor at the QuestionPro Blog recently had an excellent post with a couple of suggestions for minimizing drop-outs, which I will borrow as suggestions 1) and 2).

1) Allow Respondents to Save and Continue Later

Your respondent is halfway done with your questionnaire, and the phone and doorbell ring simultaneously.

Why not give the respondent the option to save their progress and continue taking the survey at a more convenient time? If you don’t allow this option, you’re encouraging people to close their browser and say, “oh, forget it.”

2) Set a Time Limit and Provide a Timer

If you set a time limit for the completion of a survey, along with a built-in timer, you are encouraging the respondent to give a higher level of focus on the task.

Think about it, if you are trying to finish the survey within 10 minutes, you are concentrating and working hard. You don’t want to drop out because you are in a race with yourself and you want to see if you can meet the challenge.

And now, here are three more suggestions:

3) Keep Your Survey Short

Write your survey. Cut it in half. Then cut it in half again. Then maybe, just maybe, you can add a few questions back in.

For years I’ve been hearing people talk about the importance of keeping surveys short. With the advent of mobile surveys, that advice is even more important. Yet I keep hearing horror stories about 30-minute, 40-minute or even longer questionnaires.

Stand up for your respondents by cruelly and efficiently cutting out all the fat, and maybe even some bone.

If you’re asking respondents to fill out a questionnaire, ask yourself, “Would I do this survey?” I think the answer may be revealing.

4) Provide a Progress Bar

A progress bar is a merciful thing. And it’s a smart thing. Without it, respondents are left guessing how much longer they will be completing your questionnaire.

Without a progress bar, a respondent could drop out just a few questions from the end of the survey. With a progress bar, respondents at least have a sense of the task at hand. Honesty is appreciated and goes a long way to engendering loyalty.

If your survey software offers a progress bar option, use it. If it doesn’t, consider looking for new survey software.

5) Make the Survey Convenient

If you make the survey convenient for the respondent, they will reward you with their responses and with quality data.

First and foremost, for online surveys your default assumption should be to offer the questionnaire on a traditional desktop/laptop web browser, and offer a mobile option.

Of course, this can backfire if your survey has not been properly tested and is thus terrible looking, or even completely messed up, on a mobile device. Depending on your audience, you should also consider offering offline options as well.

These are just a few options to consider for keeping your respondents “turned on” and “tuned in.” Fundamentally, it boils down to putting yourself in the respondents’ shoes and creating a questionnaire that you yourself would realistically complete.

Please share your ideas for keeping respondents engaged in your survey in the comments section below.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

This is a watershed moment in the United States, and arguably worldwide, on an issue that has been in many ways the premier civil rights question of this generation. Recent polling shows the U.S. population roughly evenly divided on the question.

I, for one, agree with Obama, and I’m glad to see he finally got some stones on this issue.

I’ve always thought, or perhaps assumed, that market researchers tend to be libertarian on social issues.

Let me test my assumption.

How much do you agree or disagree with the statement, “Same-sex couples should be able to get married.” Please respond using the MicroPoll below.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

Carol Rozwell of Gartner moderated a panel of experts on innovation in sentiment analysis:

- Leslie Barrett of TheLadders
- Bing Liu of the University of Illinois at Chicago
- Romi Mahajan of Metavana

Barrett predicted that sentiment analysis would move more toward free flowing emotion detection.

Mahajan said the two major innovations are the organizational and the philosophical. Organizationally, we will make the vast amount of social data available to all in an organization rather than the traditional “cloistered priesthoods” who control information now. Philosophically, we will move away from serial and linear information toward a constant dialectic.

Liu gave an example of how companies are using social sentiment when considering acquisition targets. He also spoke of how his wife and daughter ask him how social data is useful to them – using the example of gathering information for a mattress purchase.

Mahajan said that even if part of what we are discussing comes to fruition it will be a major change.

An audience member stated that we have not discussed today who the people are whose sentiment we are measuring, and that we need to take this information into account.

Barrett said her organization does take this type of segmentation into acccount. Liu mentioned he has analyzed gender differences. Mahajan said that looking at causation is a tall order; rather, we should be satisfied with strong associations.

Liu mentioned that sentiment analysis can make it easier to process large amounts of information; for example, who wants to read every Amazon.com comment about a product?

An audience member asked about government use of sentiment analysis data. Mahajan mentioned that it would be useful for national security purposes. Another audience member who is a professor at George Washington University in Washington, DC said he is aware that the Defense Intelligence Agency makes use of some type of sentiment analysis.

Barrett said she believes technology will not so much put people out of business but rather create additive business opportunity. Mahajan added that changes in technology lead to the demise of some types of jobs and the rise of otehrs.

Mahajan said we would be wise to plan for how to deal with “blowback” by opponents of using sentiment analysis.

Barrett said each of us should go back to our organizations and look at our data and figure out at least one thing we will do differently based on what we have learned today.

4:29pm

Ronen Feldman of Hebrew University and Digital Trowel talked about his company’s solution called Visual Care. He stated the benefits include reduced development time and increased accuracy.

4:17pm

Kevin Cocco of SproutLoop talked about the way his company uses crowdsourcing to categorize data from the Twitter and Google API feeds. He compared the crowdsourced predictions to those of the Google Prediction API.

The Google Prediction API is ppor for batch predictions and model tuning; it is good for real-time data, sccaling and easy integration. The crowdsourcing model works well when human agree; however, in this experiment they agreed only 44% of the time. Cocco concludes tweet sentiment analysis can be confusing for humans.

4:06pm

Zack Kass of CrowdFlower said that sentiment analysis has been reduced to “PNN” (positive, negative, neutral), machines can’t even accurately define PNN, and crowdsourcing provides rich feedback. He presented an analysis of Radian6 data about former U.S. presidential candidate Herman Cain which was 27% accurate according to human audit. His company’s performance of the same task had 94% accuracy, he stated.

First they ask the human coder whether the piece of data is relevant. Then they rate the sentiment polarity relative to the target. Then they code options for finer gradations of meaning.

3:53pm

Max Yankelevich of CrowdControl discussed cognitive surplus and crowdsourcing. He spoke of ideas from the book Cognitive Surplus by Clay Chirky.

We have a lot of cognitive surplus, as evidenced by the following jusxtaposed facts:

- 200 billion hours per year spent watching TV by US adults
- 100 million hours to create Wikipedia

Crowdsourcing can be thought of as “crowd computing.”

Challenges with getting things done with cognitive surplus:

- complex tasks
- accuracy rate
- lack of attention
- lack of commitment

Yankelevich advocates combining artificial intellicenge (for the computer component) and crowdsourcing (for the human element).

3:19pm

Seth McGuire of Gnip said it’s not just about Twitter. Gnip aggregates and sells social media data across multiple platforms.

McGuire asked what is the right combination of data, that is, what is the right social cocktail?

There were two key dimensions they discovered:

- Reaction time – Twitter, Facebook and Google+ were faster, while WordPress, DISQUS and IntenseDebate were slower.

- Depth – Deeper platforms were YouTube, Tumblr and Flickr; more concise networks were Twitter, Facebook and Google+

Public Relations and supply chain professionals need faster, more concise data. Product development and brand mangagement professionals need deeper, not necesarily faster data.

3:13pm

Jeff Catlin of Lexalytics gave examples of tweets his company trained its sentiment engine to “solve” which are easier for humans to understand:

- Citigroup allows leniency for victims of foreclosure
- I loveeeeeee my evo
- I have an iPhone, but I am not really feeling very happy about my iPhone
- In my opinion right now, Apple is making money on a smart marketing strategy

Here are some they have not been able to solve:

- It was awesome – for the week that it worked.
- i thought i saw a previous for that on mtv movie awards which was a joke
- I don’t get why they call it the droid incredible
- That backflip was so sick

2:50pm

Frank Cotignola of Kraft Foods said sentiment analysis is not at all ingrained in market research. He asked whether we are truly listening. It is a cultural shift in how we interact with consumers. This is a difficult change.

A big mistake some make is just listen to what is being said about brands. However, people often do not talk about brands. The better approach is to listen to what consumers are saying; then see where brands fit into the conversation.

Common objections to sentiment analysis include:

- not representative
- missing demographics
- not my consumers
- too much to read
- no time
- not what I’m used to

A way to convince people of the utility of sentiment analysis is to give examples.

One example is the ability to predict questions around the economy. Typically we look at traditional economic measures. What if we used social media to assess the economy. What is the online sentiment about things like gas prices, unemployment and food prices. You can also look at search data – for example, searches on the term “unemployment” track the unemployment rate (presumably people are looking for information about benefits).

2:43pm

Sobhan Hota of Fidelity Investments discussed how his company uses sentiment analysis of their Voice of the Customer data for direct customer outreach, identifying influential customers, and customer retention. They do coding and analysis of data that identifies the top positive and the top negative words and phrases.

2:33pm

Ryan Sager of the Wall Street journal discussed their ongoing series of sentiment analysis data presented in their weekend newspaper under the title “Sentiment Tracker: A Computational Analysis of the Conversation on Social Networks.” He gave an example of an infographic they published analyzing the reaction on Twitter and Facebook to Tim Tebow becoming a member of the New York Jets football team.

2:24pm

David Nadeau of Media Miser discussed cross-lingual media sentiment analysis. Possible solutions to the cross-lingual challenge are: creating a system for each language or applying the same system after machine translation.

They did an experiment comparing:

- English sentiment analysis on French texts
- French sentiment analysis on French texts
- English sentiment analysis on machine translated French texts

The machine translation approach worked best. Further analysis showed that combining approaches worked best.

2:19pm

Michael Tupanjanin said his company Metavana has come up with a scientific breakthrough that will wipe the slate clean. It is a break from Natural Language Processing. They apply the principles of Chaos Theory to sentiment analysis. He said their algorithm has very high accuracy and is automated.

In the past sentiment analysis has been labor intensive, with low accuracy rates and heavy in professional services. There is now an historic business opportunity because of the explosion of the social web.

2:12pm

Andera Gadeib of Dialego AG talked about her company’s process of online ideation followed by classification. Gadeib starts with the divergent – creating ideas, followed by the convergent, adding layers of complexity to the analysis.

They created an ontology of areas addressed by sentiment analysis:

- emotional
- advertising
- person
- product
- action
- location
- brand
- time
- functional

The process of divergence includes concept testing, co-creation and crowdsourcing.

Measuring emotions is important for communications, product development and more. Gadeib gave a case study for a vacuum cleaner product; they found more emotion in this space than they expected. Blogs yielded more positive emotion and engagement; Twitter had more general content and skewed more negative.

They look at sentiment focus over time in a graphic they call the “long tail.”

12:05pm

Srini Bharadwaj of RAGE Frameworks talked about his company’s provision of its “Real Time Intelligence” product to enable a major financial institution to monitor of borrowers globally. He also gave a case study of the use of the same product by a pharmaceutical company to monitor drug safety and competitive activity.

11:55am

Catherine Van Zuylen of Attensity said the growth of social media has led to renewed interest in sentiment analysis. Sentiment analysis used to be more simple. But there is a change in what is meant by sentiment analysis.

Relative sentiment: “I bought an iPhone” is positive for Apple but negative for Apple’s competitors.

Also, negative sentiment is not always bad; Sarah Palin’s sentiment ratings were very negative when she hosted the Today Show; however, the the TV ratings for that show were very high.

Compound sentiment example using a tweet about the TV show Mad Men: “I love the show but hate the misleading episode trailers.”

Another trend is ambiguity and new uses for negative words. For example: “hate” is positive when used in the phrase “I hate to see her cry.”

It is also important to take emoticons into account; and emoticons vary culturally.

It is important for your team to be on the same page with respect to the definition of sentiment analysis and its specific operationalization.

11:44am

Banafsheh Ghassemi of the American Red Cross talked about her organization’s brand – which she described as one of the most recognized in the world. Its brand value is twice that of most American non-profits. They are also leaders in mobile text donations. They are also strong in social media. They have a partnership with Dell and Radian6 to track reaction to large-scale disasters in real time. They also have a growing mobile app presence with a focus on first aid and disaster response.

The number of charities has increased by 60% in the past decade. The Red Cross is a strong brand, but it still needs to win hearts, minds and dollars. Traditional advertising such as television spots is less effective than it used to be, and in this domain, the influence of friends and relatives is more influential. Influencers have big megaphones via social media, and they have the multiplier effect on their side. Ghassemi mentioned the recent Susan G. Komen controversy as a negative example of the effect of this multipler effect on a charitable organization.

The American Red Cross cares about the experience people have at touch points, change in sentiment, and executive visibility to systemic issues and investment prioritization.

It’s not just Twitter and Facebook. Yelp, for example, has user feedback on blood-giving touchpoints.

Advantages of analyzing social data are:

- real-time feedback
- it is a leading indicator
- competitive intelligence
- best practices

Opportunities with social data include:

- outreach (particularly youth and minorities)
- new policies
- product ideation
- process ideation

The death of the survey is overrated. Surveys give the American Red Cross lots of detailed feedback.

Beware of channel bias – different data sources tend to yield different flavors of data.

Be segment-appropriate. “Red Bull is not Red Cross.”

Beward of “Google Translate Syndrome” – sentiment platforms can lead to machine-applied incorrect information. In a recent disaster response, only 26% of positive comments were coded as positive in a sentiment platform (as compared to live coders).

Take a balanced approach, and do not lose sight of your traditional channels as you explore new ones.
11:15am

Chris Frank of American Express and Paul Mangone of Opnet Telecom are the authors of “Drinking from the Fire Hose.” They discussed their approach to online sentiment.

They apply the concept of the election “swing voter” to that of sentiment. Who are the people with neutral sentiment, who have the opportunity to move either in a positive or a negative direction. Which are the neutrals that lean in either a positive or a negative direction.

Frank and Mangone outlined a taxonomy of increasing involvement with a brand online. The steps, in order, are:

- Like it
- Know it
- Buy it
- Advocacy

Influence = power x platform

Power is derived in three ways:

- positional
- expert
- informational

They showed an “influence map” with platform (relevance, reach and amplitude) on the y axis and power (positional, expert, informational) on the x axis.

10:17am

Richard Brown of Thomson Reuters discussed his company’s provision of “news analytics” to financial markets in order to predict equity movements.

They provide 82 fields of data on all manner of financial news items, including:
- time stamp
- company identifier
- attribute
- type
- genre
- headline
- relevance
- sentiment
- degree of positive, negative or neutral
- first mention of the company
- topic codes

They are also coming out with something called Market Response Indicators which apply machine learning to determine which of the 82 fields are the most important at the stock, market and sector levels.

Thomson Reuters is now plugging in social media data. Markets are now more automated, and traditional analysts are doing what quants used to do.

You have to have big data in your business plan – forget it if you don’t.

Thomson Reuters is planning to take the problem of big data and turn it into an opportunity. They compare signals from internet news and social media output to signals generated from premium news (Reuters).

9:52am

Carol Haney of Toluna described text analysis as looking for the right needles in the haystack. She also noted that much of the data is negative in nature.

There is quite a lot of noise when selecting the data to analyze. It is important to gain an understanding of whether particular information is applicable.

Planning up front is important when embarking on an analysis. The steps are:

- plan your analysis
- harvest the data
- structure and understand the data
- validate the data with a quantitative survey

Haney noted it is important to weight to census rep and use a quality panel. Also, where to scrape depends on where you are in the world.

She presented a case study about Victoria’s Secret’s Dream Angels and Pink brands. Data were harvested from Facebook, Twitter and blogs. Clustering was used to identify and remove promotions. Then a classification scheme of brand and style was created.

Data were very domain specific. For example, the word “ass” less negative in this context because the product is underwear.

Haney also only looked at stronger setiment.

Issues identified from the analysis were thus:

- 2% said stores are not carrying the right size in swimwear
- 7% said Victoria’s Secret isn’t addressing the needs of women outside the 18-24 age group
- 1% said Victoria’s Secret merchandise is ugly

Haney then validated the comments about carrying the right size by conducting a survey of VS customers. Seventeen percent agreed about need to carry bigger sizes in store.

9:22am

Professor Jan Wiebe of the University of Pittsburgh described the process of “supervised machine learning,” as part of Natural Language Processing.

In this process there is a set of training data which is analyzed to create a learning algorithm. That algorithm is then applied to a set of data for which predictions are made for labeling the text sentiment.

Disadvantages to supervised machine learning are that it is expensive and time consuming to create training data expensive and time consuming.

Further, the meanings of words are domain dependent. Performance of machine learnning suffers when training and test data come frome different domains. Cross-domain sentiment analysis methods can help increase accuracy when analyzing data across domains.

Wiebe also discussed “sense level processing.” Senses are different meanings of words depending on context. Many words have multiple senses – for example, “interest,” “alarm” and “trust.” Further some senses of words are opinion-bearing while others are non-opinion bearing. When analyzing sentiment, non-opinion bearing senses are false hits. Data show that simply analyzing opinion polarity rather than each specific sense of a word can lead to higher accuracy.

Wiebe also described data acquisition, including the use of data annotators through the Amazon Mechanical Turk (AMT) service of Amazon.com. Data show that expert annotators perform better over time than those contracted through AMT.

“Active learning” is a process that can be used to reduce the amount of training data needed to train reliable systems. The most informative, least redundant data are analyzed first, then les efficient data are analyzed, and the analyst iterates through more data until satisfied.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

The agency has been led since 2009 by a highly respected academic and leader in the field of survey methodology and social research, Dr. Robert Groves. (Anyone who knows Dr. Groves or his work was appalled by the politicization of his nomination as Director in April 2009.)

It is not surprising, then, that nearly all of the bureau’s survey work is done by mail, telephone, and – gasp! – in person. Right?

Well, that’s changing.

Dr. Groves is already making plans for the Bureau to offer an online option in the 2020 decennial census.

OK, but 2020 is a long way off.

But even sooner, there are recently announced plans to start using an online option by January in the American Community Survey (ACS), the ongoing, highly detailed, mandatory study which is the replacement for the former “long form” in the decennial census. The results of the ACS online option will inform plans for Census 2020.

The ACS plans, which are not yet approved, call for respondents to be asked first to respond online; those who have not responded online will receive a paper questionnaire after about 2 weeks, followed by a postcard reminder.

The Bureau estimates about half of those currently responding by mail in the first month will respond online, representing a significant savings in the cost of mail data collection.

You can be sure the Bureau will be publishing detailed reports on its ongoing tests of online survey methods. Those tests are sure to yield lessons valuable to market researchers. Let’s face it, we don’t have the time, the budget or the scale to do tests comparing and contrasting data collection modes on a large scale.

Now if we could just get them interested in mobile data collection… .

Please share your thoughts on the U.S. Census Bureau’s plans for online data collection in the comments section below.

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess

Check out Research Access on Tuesday for live updates.

Better yet, join me at the Symposium at Lighthouse International in Manhattan.  You can still register for $100 off using the code FOAF.

Here are some of the sessions I’m most looking forward to:

• “Tween Pants Cut Too Low!! (or, Combine Survey Research & Social Monitoring to Discover the Unknown)” by Carol Haney, Toluna
• “Emotional Versus Rational in Customer Decision Making” by Chris Frank, American Express, and Paul Magnone, Openet Telecom
• “Real Time Intelligence Solutions,” by Srini Bharadwaj, RAGE Frameworks
• “Political Sentiment Analysis,” by Dr. Stuart W. Shulman, Texifter
• “Market Research Beyond Sentiment: Differentiating the Engaged and Pleased,” by Andera Gadeib, CEO, Dialego AG
• “Sentiment As A Service,” by Michael Tupanjanin, Metavana
• “Capturing Sentiment via Customer Intelligence,” by Sobhan Hota, Fidelity Investments
• “’How Can I Listen If I’m Talking?’: The Power Of Social Media Listening,” by Frank Cotignola, Kraft Foods

I hope to see you in New York or online on Tuesday!

© Research Access, 2012. | Permalink | Add a comment | Add to del.icio.us | Share on Facebook
Follow ResearchAccess on Twitter @ResearchAccess