Affectiva: Measuring Emotion Over the Web

I just read about a new company called Affectiva, which "grew out of collaborative research at the MIT Media Lab to help people better measure and communicate emotion."

This blog has previously discussed how human behaviour and sentiment is being analysed using data from the web:
(i) Web search volume can can improve forecasts of the current level of activity for a number of different economic time series, including automobile sales, home sales, retail sales, and travel behaviour (Google research paper on "predicting the present").
(ii) What people are searching for today can be predictive of what they will do in the near future (Yahoo! PNAS paper)
(iii) There have been various initiatives to make sense of sentiment data from Twitter
(iv) Facebook are analysing a rich seam of social networking and affect data
(v) Dan Hopkins and Gary King have developed an automated program to gather sentiment from online blogs

What makes Affectiva different to all of the above, is that (through their project Affdex), they have provided a way to collect web-data on physically expressed human emotions, not stated mood or revealed (search) preferences. Data on facial expressions and gestures — while people interact naturally with media, or other online experience — can be collected using Affdex. One could argue that this is closer to the objective truth than self-reported affect or observed behaviour.

Also, Affectiva's Q Sensor project measures galvanic skin response using a wearable, wireless biosensor. The sensor also measures temperature and activity.

Bundle: Microeconomic Insights from Citibank Data

ReadWriteWeb.com: "Thanks to a cooperation with Citibank and other third-party data suppliers, Bundle is able to compile detailed statistics about how Americans are spending their money. While lots of banks also compile this data, Bundle is the first service to make this data easily accessible." Using Bundle, American citizens can investigate whether their spending patterns are in line with that of other people in their age and income group in their neighborhood. This brings a new twist to ideas from behavioural economics such as reference-dependency and other-regarding preferences.

Bundle plans to update its data on a quarterly basis. It also plans to allow users to enter their exact spending habits by either entering the data by hand or by giving users the option to upload credit card statements directly. In relation to self-reported spending habits, Dave, Liam and Colm have done research on "Experimental Tests of Survey Responses to Expenditure Questions". Bundle is also set to provide budgeting advice based on its users' spending habits. Bundle is a joint collaboration between Microsoft, Citigroup and Morningstar. The site fully integrates with Facebook and Twitter. A brief video from the folks at Bundle.com is shown below.

Data Entry Operator Job offer from Unitcom International Ltd.

Data Entry Operator    

No. of Vacancies :  40






Job Description / Responsibility

    * Should be skill on data entry.
    * Good typing speeds are preferable.
    * Good knowledge in internet browsing.
    * Teamwork and excellent interpersonal skills highly required.
    * Must have good etiquette and politeness.
    * Person should be smart and energetic


Educational Requirements

      S.S.C or H.S.C or University or National University Student



Additional Job Requirements
    * Age 18 to 30 year(s)
    * Male or Female can apply
    * The applicant must possess Smartness and positive attitude
    * The applicant must have extrovert & dynamic in nature
    * Well efficient in Computer application
    * Must be excellent in English Communication
    * Able to work under stress & in night shift
    * Should be able to work in a team under pressure with multiple functions

Salary Range

    * Negotiable

Other Benefits
      In-house training and personnel development will be provided

Job Location

Dhaka

Job Source

Bdjobs.com Online job posting

Apply Instruction

Send your CV to hr.admin@wclbd.com

or

Special Instruction : If your candidacy matched with our requirements, please send your application along with updated Resume and one recent passport size color photograph to Unitcon International Ltd., A Partner of Wave Consortium Ltd., 7th floor, 12 BDBL Bhaban, Kawranbazar, Dhaka-1215. Phone: +02 818 9365, Cell: 01673 91 50 44, 01752 12 38 15.

Applicant must enclose his/her Photograph with CV.
Application Deadline: February 09, 2011
Company Information
Unitcom International Ltd.
Address : 7th Floor, 12 BDBL Bhaban
Web : www.wclbd.com
Business : Deta Entry

Tesco Metrics: Every Little Bit of Data Helps

Liam linked to an article in the Guardian earlier this week, which was all about Nudge. One comment in the article was that "while shopping, working, or even deciding on who to share their lives with, individuals are less thoughtful and less calculating than modern-day economists... typically assume." This blog-post zones in on shopping, in particular the data-analysis of consumer purchasing behaviour at Tesco. The Guardian article linked above also suggests that "any critic who points out that that's hardly news to the women...(and) the men at Tesco... is spot on." Indeed, Tesco have been conducting interesting micro-level analysis on individual behaviour for many years now.

An informative article on this topic was written by Jenny Davies in the Sunday Times last year. According to Davies, Tesco gets its data from its loyalty clubcard scheme; this was launched 15 years ago with much fanfare - the advert below may jog memories for some readers. Davies also informs us that around this time last year, Tesco was tracking "the shopping habits of 16 million families across Britain, delivering an extraordinary insight into their lives — not only for itself but for companies such as Coca-Cola, Nestlé and Unilever, which buy the rights to the data." Readers in the Republic of Ireland might also remember that the Tesco Clubcard was launched there on the 13th. Oct 1997. To date almost 800,000 members have joined in the Republic.



Jenny Davies also tells us that: "Each bill detailing every item in a customer’s shopping basket is logged in a data centre in London Docklands and decoded by Dunnhumby, the marketing firm that is in charge of the scheme. It has to process 100 baskets a second — six million transactions a day. This helps Tesco to decide which products should go on to the shelves at what times, and in early trials it increased sales by as much as 12% in some of the supermarkets." According to the Guardian (in this article), the power of the clubcard was demonstrated in 2009, "when Tesco harnessed the card's database to halt the exodus of shoppers to cheaper retailers because (of) the recession, by doubling the points available to shoppers."

In a blog-post on Tesco data from two years ago, Tony Hirst desribes the early analysis conducted by Dunnhumby, and how this has changed over the last 15 years. A couple of months ago, Dunnhumby (and its recently departed co-founders) were profiled in the Guardian. The article says:

According to company lore, there was a 30-second silence after Humby presented the initial trial's results to the Tesco board, until the then chairman, Lord MacLaurin, declared: "What scares me is that you know more about my customers after three months than I know after 30 years."

One question that readers might have is: what's in it for club-card holders? According to Tony Hirst, a good place to get an answer to this question is the book: Scoring Points: How Tesco Continues to Win Customer Loyalty. Hirst describes the "Clubcard customer contract: more data means better segmentation, means more targeted/personalised services, means better profiling. In short, the more you shop with us, the more benefit you will accrue." According to the Marketing Week magazine, "from the day of its launch in February 1995 the Tesco Clubcard was immediately embraced by customers attracted to the 1% discount off their shopping bills. But its long term success has not been built on discounts alone, rather on the personalisation of the shopping experience."

However, perhaps the last word should go to UCD social psychologist Ken McKenzie, writing on his A Head in Business Blog: "I don’t have a loyalty card, and every time I’m in Boots, Tesco or Dunnes, and they ask if I have one, I feel a slight sense that I should justify why I don’t, as it it’s odd to not have one. And according to rational actor theory in Economics, it is odd to not have a loyalty card and avail of discounts. However, there’s a small but growing body of work in the overlapping area between Psychology and Economics that might explain why (some) people might behave like me."

Economics at Yahoo!

I've blogged before about Hal Varian, Chief Economist at Google, and how his role in the organisation has been central to Google's business model. I have also mentioned pioneering work at Yahoo! Research on the effectiveness of online advertising, the battle between Google, Yahoo! and Bing in the economics of internet search, and the statement by Yahoo! Research that they routinely compete for talent with the top ten economics departments in the world.

However, it has only now come to my attention that Varian has a counterpart at Yahoo! The economist in question is Preston McAfee, on leave from the position of J. Stanley Johnson Professor of Business, Economics, and Management at the California Institute of Technology. At Yahoo!, Professor McAfee is Vice President and Research Fellow at Yahoo! Research in Burbank, CA, where he leads a group focused on microeconomics research. Here is a list of recent publications from the Microeconomics and Social Systems cluster at Yahoo!

Professor McAfee wrote Introduction to Economic Analysis, a free, open-source text that spans both principles and intermediate microeconomics. In 1994, the FCC in the USA auctioned access to a number of radio frequencies for new communications services, using an auction designed by Paul Milgrom, Robert B. Wilson, and Professor McAfee, and raised over $17 billion. This auction design was copied around the world. McAfee, Milgrom, Wilson and John McMillan (1951-2007) formed a company, Market Design, Inc., that advises governments on how to maximize the return from sales of radio frequencies, mineral rights, airports, and other assets.

Returning to Yahoo!, The Register magazine describe the company's Right Media exchange — a display advertising marketplace that matches advertisers with publishers and ad networks — as (by one measure) the largest exchange in the world, running over nine billion auctions each day. Finally, the Yahoo! Advertising Blog is also an interesting read; the current post - Mad Men No More - features a discussion by "advertising’s new guard" on how they are re-defining the industry.

Links of Interest: 29th September

1. President Barack Obama chose Austan Goolsbee to succeed Christina Romer as the head of the U.S. Council of Economic Advisers. Here, the Wall Street Journal do a profile of Goolsbee.

2. The Guardian: a "nudge unit" set up by David Cameron in the Cabinet Office is working on how to use behavioural economics and market signals to persuade citizens to behave in a more socially integrated way.

3. The Daily Telegraph on Rory Sutherland's quiet behavioural economics revolution in the advertising industry.

4. Greatest Good: "a unique firm formed with the goal of applying rigorous, cutting-edge data analysis and economic methods to the most salient problems of business and philanthropy." Founding partners include Steven Levitt, Gary Becker, Daniel Kahneman and John List. Affiliates include David Laibson, Emily Oster, Steven Pinker and Richard Thaler.

5. The U.S. National Commission on Fiscal Responsibility and Reform. They have a separate mandate to the Congressional Budget Office. "The Commission is charged with identifying policies to improve the fiscal situation in the medium term and to achieve fiscal sustainability over the long run."

6. A fascinating read for any Ph.D. student in Economics, or Ph.D. economist: 'Market Structure in the Production of Economics Ph.D.s'. Frank A. Scott, Jr. and Jeffrey D. Anstine; Southern Economic Journal Vol. 64, No. 1 (Jul., 1997), pp. 307-320.

7. The (Irish) Department of Education and Skills Inventory of Data Sources: "This document contains a matrix of educational data sources which are available from the Department of Education and Science and the agencies under its aegis."

8. University Attendance Scanners: "Northern Arizona University has installed electronic devices that record student attendance in an effort to boost freshmen grades and lift lagging graduation rates. But some students say the monitoring makes them feel less independent." (Southern California Public Radio).

9. "The Production and Deployment of an On-line Video Learning Bank in a Skills Training Environment" - Gerald Cannon, Mary Kelly, Colette Lyng, Mary McGrath; AISHE-J: The All Ireland Journal of Teaching and Learning in Higher Education, Vol 1, No 1 (2009).

10. For economics undergraduates: the Irish Taxation Institute Fantasy Budget Competition. Who needs fantasy football?

Some Online Irish Education Data That You May Not Have Seen

1. Primary-school class sizes for the last four years: data available by primary-school and class-room. The source for this data is the Annual Census of Primary Schools.

2. Retention Rates of Pupils in Second-Level Schools: 1991 to 2001 Entry Cohorts. These data are aggregated but provided on a county by county basis. The overall Leaving Certificate retention rate has increased in the five years between the 1996 cohort and the 2001 cohort: from 81.3% to 84.7%.

3. The State Examinations Commission website has a statistics page providing Leaving Certificate and Junior Certificate grade information for every subject, and by gender, for the years 2000-2010.

Washington Institute for Health Metrics and Evaluation Datasets

Students and researchers should keep a close eye on the evolving data infrastructure at the Institute for Health Metrics and Evaluation at University of Washington. The Maternal Mortality data page here is an example. Many of these data-sets were constructed for large scale evaluations of public health initiatives developed with a view to the Millenium Development Goals. They are very useful resources also for development economists and do not yet seem to have been widely used in this area. They are also, for the most part, downloadable for free in a "one-click" excel format. The growing publication list of this Institute is also worth looking at carefully.

California Knocks Down Data Firewall

California is adept at building firebreaks to stop advancing wildfires throughout the state. The inferno that is the student-teacher firewall issue was apparently doused yesterday when Governor Arnold Schwarzenegger signed a bill that eliminates a statutory ban on using student achievement data to evaluate teachers. The existence of such a restriction would have deemed California ineligible for a federal Race to the Top competitive grant award.

Here is the Governor's press release.

Here, here and here are background posts on the student-teacher data firewall issue in California.

Is California's "Firewall" Penetrable?

California Superintendent of Public Instruction Jack O'Connell countered criticism by Education Secretary Arne Duncan about a state law restricting the use of student assessment data in teacher evaluations. As reported in today's Los Angeles Times, O'Connell highlighted Long Beach Unified as a school district that does exactly that.

California's top education official sought Tuesday to counter federal criticism of the state's reluctance to use student test scores to evaluate teachers, paying a visit to Long Beach to highlight one of the few California school districts to make extensive use of such data.

The Long Beach Unified School District's use of student scores to assess the effectiveness of programs, instructional strategies and teachers is a rarity in California, and state Supt. of Public Instruction Jack O'Connell called it a model for other California school districts during a hastily arranged round-table discussion.

At issue is a 2006 California law that prohibits use of student data to evaluate teachers at the state level. O'Connell said Obama and Duncan misunderstand the law, which does not bar local districts from using the information.

O'Connell also released a statement on this issue last week.

Long Beach Unified is a 2009 finalist for the Broad Prize and was recently profiled by TIME magazine as one of the top urban school systems in the nation.

(Re)Focusing on What Matters

Last week I spoke at a meeting of the Lumina Foundation’s Achieving the Dream Initiative, a meeting of policymakers from 15 states all working to improve the effectiveness of community colleges. At one point, a data working group shared results of its efforts to create new ways to measure college outputs. This was basically a new kind of report card, one capable of reporting results for different subgroups of students, and enabling comparisons of outcomes across colleges. Something like it might someday replace the data collection currently part of the IPEDS.

While it's always gratifying to see state policymakers engaging with data and thinking about how to use it in meaningful ways, I couldn’t help but feel that even this seemingly forward-thinking group was tending toward the status quo. The way we measure and report college outputs right now consistently reinforces a particular way of thinking-- a framework that focuses squarely on colleges and their successes or failures.

What’s the matter with that, you’re probably wondering? After all, aren’t schools the ones we need to hold accountable for outcomes and improved performance? Well, perhaps. But what we’re purportedly really interested in—or what we should be interested in—is students, and their successes or failures. If that's the case, then students, rather than colleges, need to be at the very center of our thinking and policymaking. Right now this isn't the case.

Let’s play this out a bit more. Current efforts are afoot to find ways to measure college outcomes that make more colleges comfortable with measurement and accountability--and thus help bring them onboard. That typically means using measures that allow even the lowest-achieving colleges at least a viable opportunity for success, and using measures colleges feels are meaningful, related to what they think they’re supposed to be doing. An example: the 3-year associates degree completion rates of full-time community college entrant deemed “college ready” by a standardized test. We can measure this for different schools and report the results. Where does that get us? We can then see which colleges have higher rates, and which have lower ones.

But then what? Can we then conclude some colleges are doing a better job than others? Frankly, no. It’s quite possible that higher rates at some colleges are attributable to student characteristics or contextual characteristics outside an individual college (e.g. proximity to other colleges, local labor market, region, etc) that explain the differences. But that’s hard to get people to focus on when what’s simplest to see are differences between colleges.

It's not clear that this approach actually helps students. What if, instead, states reported outcomes for specified groups of students without disaggregating by college? How might the policy conversation change? Well, for example, a state could see a glaring statewide gap in college completion among majority and minority students. It would then (hopefully) move to the next step of looking for sources of the problem—likely trying to identify the areas with the greatest influence, and the areas with the most policy-amenable areas of influence. This might lead analysts back to the colleges in the state to look for poor or weak performers, but it might instead lead them to aspects of k-12 preparation, state financial aid policy, the organizational structure of the higher education system, etc. The point is that in order to help students, states would need to do more than simply point to colleges and work to inspire them to change. They’d be forced to try and pinpoint the source(s) of the problems and then work on them. I expect the approaches would need to vary by state.

Don’t get me wrong, I’m not trying to absolve any college of responsibility for educating its students. What I’m suggesting is that we think hard about why the emphasis right now rests so heavily on relative college performance—an approach that embraces and even creates more insitutional competition—rather than on finding efficient and effective ways to increase the success of our students. Are we over-utilizing approaches, often adopted without much critical thought, that reify and perpetuate our past mistakes? I think so.

Image Credit: www.openjarmedia.com

Obama's Firewall

Michele McNeil at Education Week's Politics K-12 blog reports that President Obama himself approved the Race To The Top provision that restricts these competitive grants for going to states that restrict the use of student achievement data in teacher evaluations. Short of statutory or regulatory changes made by these 3 states, should we be saying au revoir to the Golden, Empire and Badger states?

Only two things can render a state ineligible for Race to the Top grants.

And only one of them is a biggie: the student-teacher data firewall issue.

This effectively means New York, California, and Wisconsin, at the very least, are ineligible for Race to the Top—or will at least have some explaining to do. They have laws on the book that essentially bar the use of student-achievement data in some teacher-evaluation decisions.

Erin Richards at the Milwaukee Journal-Sentinel picks up the story, too.

Background here and here.

Cheerleading for NCLB

I guess my reaction to today's Washington Post shout out to No Child Left Behind ("'No Child' in Action") from former U.S. Education Secretary Margaret Spellings is a question: "If NCLB's accountability alone is such a silver bullet, then how come test scores at the high school level didn't improve?"

Although Spellings mentions that NCLB requires math and reading tests in grades 3-8, it is quite disingenuous of her not to mention that such tests were also required in high school. If the achievement gains aren't sustained through high school, what real difference does it make?

The wise Aaron Pallas offers his take on this issue ("Wishful Thinking"), calling into question Spellings's claims:

But what portion of those trends can be attributed to NCLB? Margaret Spellings refers to changes since 1999, which is convenient for her story, because there were sharp increases in grade 4 reading between 2000 and 2002, and in grade 4 and grade 8 math between 2000 and 2003. But NCLB was signed into law in January, 2002; the first final regulations dealing with assessment were issued in December, 2002; and initial state accountability plans were approved by the U.S. Department of Education no later than June, 2003. The 2003 main NAEP was administered between January and March of 2003. Is it realistic to claim that NCLB affected scores before the 2003 NAEP administration? I, and a great many other analysts, think not.

Only in Margaret Spellings’ world can NCLB affect NAEP scores for the four years before the law was passed and implemented. Now that’s wishful thinking.

UPDATE -- Diane Ravitch comes to similar conclusions in her blog post.

Thus, when one looks at the patterns, it suggests the following: First, our students are making gains, though not among 17-year-olds. Second, the gains they have made since NCLB are smaller than the gains they made in the years preceding NCLB. Third, even when they are significant, the gains are small. Fourth, the Long Term Trend data are not a resounding endorsement of NCLB. If anything, the slowing of the rate of progress suggests that NCLB is not a powerful instrument to improve student performance.

Caveat emptor.

A Blank Prescription for Policy Reform

Stanford has yet another interesting publication, Pathways, which integrates critical analysis of research on poverty with bold prescriptions for policy reform. I'm a bit late to the game in mentioning this, since the first issue came out in December, but then again as Mom-to-toddler a 6- month lag or so on my reading ain't bad.

I'm especially a fan of Becky Blank's piece which includes a list of priority efforts for antipoverty programs. Blank is whip-smart, and was recently a contender for the chancellorship here at Madison. Darn it all, she wasn't chosen.

Here's a quote from near the end of her article, which should illustrate why many of us would've loved to have her here. "Social policy evaluation is one of the least well appreciated tools of long-term policy design."

If there was one message I'd send to the presidential candidates, it's that when tackling ANY area of policy reform, please please please take the evaluation of your programs seriously. No, data isn't entirely objective, but it's a whole heck of a lot more objective than simply deciding something's working -- or not-- based entirely on politics or ideology.