We’ve seen a lot of new visitors to the site lately, and even better – many first-time contributors.  Welcome!

We thought this would be a good opportunity to mention a few ways to become familiar with the ins and outs of content contribution on Hunch:

• The #1 best way to get a sense of what makes a good Hunch topic is to ‘play’ a bunch of existing topics.  Take note of how questions are formulated and how result titles and descriptions are formulated.  A great deal of what we’d call a ‘confused contribution’ originates from new users who have played zero existing topics so far.  Summary: Walk around the party and shake a few hands before you jump up on the food table and start belting out your solo performance.
• Before you create a new topic, search for your idea in Hunch’s search box (top right).  We continue to see a lot of exact topic duplicates, and generally these never make it out of the workshop.  If you find that the topic already exists on Hunch, play it and then add new (or edit existing) results or questions if you feel it could be improved.
• Hunch is for recommendations.  Hunch is not a polling site, an answers site, or a search engine.  Anything that can be answered with a straightforward, single, factual answer doesn’t belong on Hunch. To answer that kind of stuff, we would direct you to your favorite search engine or answers site, which is exactly what we do for those types of queries. We elaborate on this point a bit in a former blog post.
• Have a look through Hunch’s style guidelines to learn about the format and structure for Hunch topics, follow-up questions, and results.  To cite just one example, follow-up questions should be formulated in the second person (”Do you prefer this or that?”) while their answers should be formulated in the first person (”I prefer A, or A is better for me“)
• There’s also good info in the FAQs that can help you if you’re feeling befuddled.

Now, for a slight guideline change which may be new news for veteran contributors: when possible, it’s preferable that Hunch topic names be formulated in a way which describes their results, *not* using a question formulation. So, a topic should be called “New Cars” (not “Which new car should I buy?”) or “NYT Bestsellers” (not “Which NYT Bestseller should I read?”)  We’ll be gradually converting existing topic names from question form to “description” form in cases where it’s easy to do so.

The “description form” isn’t always possible, though, particular for “either/or” or “yes/no” topics.  (e.g.: Should I be my own general contractor?)  So it’s ok to use a question formulation when there doesn’t seem to be a better way or when it’s necessary for clarity.

Keep all the great contributions coming.

We’re very psyched to announce that we’ve closed a $12 million Series B funding led by Gideon Yu at Khosla Ventures.  Gideon held senior finance roles at Yahoo, YouTube, and Facebook, and is an all-around nice guy who also happens to be incredibly smart.  He’s joining Hunch’s Board of Directors and we look forward to tapping into his experience and guidance.

As Caterina notes in this blog post, Hunch’s goal is to build a “taste graph” of the web, connecting every person on the web with every entity on the web (whether it be a camera, hotel, author, or anything else).  This new funding will help us accelerate progress towards that goal.  You’ll also be seeing a lot of great new features very soon.

Thanks to all of our passionate site users and contributors who have helped Hunch progress so far in less than a year since our public launch in June 2009!

Today we’re releasing a new report called Mommy Wars: Real Issue or Media Myth?

The report first divides mothers into several distinct groups according to how they answered various “Teach Hunch About You” questions, with subgroups defined as “Mainstream vs. Unique” moms, “Traditional vs Progressive” moms, and “Straight Arrow vs. Rule Bender” moms.  Then it explores whether there are differences in the respective groups according to attitudes about family, education, and parenting.  You can read the full report here (or download pdf version here.)

Background – Mommy Wars in the Media

From CNN to Motherlode to Momversation, the subject of “Mommy Wars” has been one of contentious debate. When Leslie Morgan Steiner first published Mommy Wars in 1986, she presented the inner turmoil felt by mothers when making and defending their personal decisions either to stay at home with their children or to pursue their careers. The challenges and joys of being a modern mom are always a fertile topic of discussion on parenting and mom blogs.

In the past twenty-five years the term “Mommy Wars” has become a blanket name for conflicts between mothers making different parenting choices, ranging from breastfeeding to sleep training. In many cases these quarrels are presented as anecdotal experiences, without much data on attitudinal differences between proponents of different parenting styles.

In this study we’ve approached the “Mommy Wars” phenomenon from an attitudinal perspective, examining differences in family values, educational philosophies, and parenting practices within different types of mothers. The question is whether mothers with distinct attitudes towards uniqueness, generational standards, and rules will exhibit similar distinctions regarding their attitudes about family, education, and parenting.

Key findings

• Mainstream moms tend to be more insular and value family in traditional ways. Unique moms have more liberal views, are more likely to give their children original names, and put more value on “giving back.”
• Traditional moms are more apt to approve of alternative educational approaches (teacher merit pay, homeschooling) to maintain student standards, but are less likely to care for contemporary content (such as Harry Potter). Progressive moms are more prolific readers and more likely to support affirmative action programs.
• Rule bending behavior in moms has little correlation to either childhood experience or parenting philosophy. In other words, just because a mom was a little on the wild side growing up doesn’t mean she’s more permissive or uninvolved as a parent.
• Mommyhood trumps all: Despite the presence of some differences among the sub-groups we explored, there are many areas where both groups are very similar, suggesting that a mom’s basic instincts to nurture and protect her children are fairly independent of some of the unique personality traits of the mom herself.

There are a lot more details and insights in the full report.  Check it out!

The issue about whether gays and lesbians should be allowed to serve in the military is a hotly debated topic in the United States.

There are actually 3 main sides to the issue: 1) whether to allow openly gay men and women to serve, 2) whether gays and lesbians should be completely barred from serving (while this isn’t being considered, it’s still an opinion held by plenty of people), or 3) whether a “Don’t Ask/Don’t Tell” policy (which has been in place since 1993) should continue.  There’s plenty of debate happening about all angles of the issue, even as initial steps are put in place to prepare for the possible repeal of the current DADT policy.

We looked at some of the aggregate data from Hunch users to see how they felt about this issue.  Note that we included all Hunch users (not just those in the US) who had answered the following “Teach Hunch About You” (THAY) question:

Overall opinions

Among more than 45,000 Hunch users who answered the question, 72% answered that gays should be allowed in the military, 24% favored keeping “Don’t Ask, Don’t Tell”, and 4% thought that gays shouldn’t be allowed to serve at all.

The 72% answering that gays should be allowed to serve represents a higher percentage than the 54% of Americans estimated to favor repeal of DADT.  There are at least 2 main reasons to explain the difference:

1) Hunch’s current audience skews towards young adults (18-34), technophiles, and liberals.  These groups tend to be more accepting of a policy allowing openly gay men and women to serve, as the charts below illustrate.

2) Hunch’s question framed the issue with 3 possible outcomes while other polls simply ask about keeping or repealing DADT.  This likely has an effect on how people answer.

Partially mitigating these factors is the fact that Hunch’s audience is global, and Hunch users outside of North America tend to be less tolerant of allowing gays to serve in the military.

We haven’t manipulated the Hunch data to re-weight or adjust for these differences.  Rather than go through all those shenanigans, we figured we’d present the unaltered data as-is, because it’s still interesting to explore the relative differences in opinions among different sub-groups.

Opinions by age of respondent

Hunch users 18-49 have the highest incidence of answering that gays should be allowed to serve; this support drops among older and younger people.

Opinions by political ideology

Not surprisingly, conservatives have the highest rate (46%) of being in favor of keeping the current DADT policy in place.  A further 14% of conservatives favor completely barring gays and lesbians from serving.  Respondents who are “middle of the road” show significantly higher rates of acceptance, with liberals more tolerant still.

Opinions by comfort with technology

The minority of Hunch users who say they are “not comfortable with technology” show a greater resistance to gays in the military, with 36% of these respondents indicating that gays/lesbians should either be barred or allowed to serve only within the context of a Don’t Ask Don’t Tell policy. (as you might expect, this measure is also highly correlated with age)

Opinions by gender of the respondent

Females (currently about 35% of Hunch users) are significantly more likely than males to favor allowing gays or lesbians to serve; male Hunch users are 37% more likely to favor keeping DADT than females are.

Opinions by continent of residence

Support for allowing gays in the military is highest among Hunch users residing in North America and Oceana.  Support falls significantly among users residing in Asia and Africa, with a quarter of Hunch users from Africa feeling that gays should be completely banned from the military.  It’s interesting that European Hunch users are more conservative on this issue than North American Hunch users, since many European countries already have policies in place allowing gays and lesbians to openly serve.

Opinions by type of military service

Among Hunch users who have ever served in the military, roughly 1/3 favor a DADT-like policy.  54% of former officers and 57% of former enlisted people favor a completely open policy, while 11% of both groups favor a complete ban.  The military’s own poll among active troops indicates about a 30% rate of favoring repeal of DADT, with 51% opposed to a repeal.

Opinions by whether respondent has gay friends

Having a close friend who is gay or lesbian is a significant predictor for someone’s opinions about gays in the military.  The 59% of Hunch users who say they have a close gay or lesbian friend are 41% more likely to favor allowing gays to serve.  A third of Hunch users with no close gay/lesbian friends favor keeping DADT.

The data used in the charts above is based on Teach Hunch About You (THAY) questions which were answered by Hunch users between March 2009 and February 2010.  You can check out more Hunch reports on many other topics here.

Ready or not, February 14^th is right around the corner.  So today we’re releasing a new report called  “Bitters” vs “Lovers”: What our Valentine’s Day outlook says about us.  You can see the full report here.

The report looks at the similarities and differences between those who say they love Valentine’s Day and those who confess that they’re “bitter” about it. We looked at four main areas of their lives: basic demographics; experiences and attitudes about love, sex, and relationships; attitudes about themselves and relating to other people; and personal beliefs.

Here’s a summary of what we found:

Demographics

• Men are more likely to be bitter about Valentine’s Day than women. Women are more likely to love the holiday.
• People who love Valentine’s are more likely to be in a long-term relationship and have kids. (so who says that settling down and having kids has to kill romance?)

Love and relationships

• People who are bitter about Valentine’s Day are more likely to feel that they have more difficulty with relationships than the average person their age. They’re also skeptics when it comes to the concepts of soul mates and love at first sight, and they’re less likely to be interested in talking about feelings and relationships.
• People who love Valentine’s Day are more likely to make romantic gestures such as sending flowers, engaging in PDA, and sending cards. They’re also more likely to forgive a cheating partner and more likely to bring up the state of the relationship with their partner in the first place.

Self and relating to others

• A person who is bitter about Valentine’s is more likely to say their IQ is higher than their EQ (emotional quotient), is more likely to say they don’t like themselves most of the time, and is more likely to feel that they don’t know who they are. He or she is more likely to identify as an introvert and is more likely to focus on their career.
• Someone who loves Valentine’s Day is more likely to identify as a “people person” and focus on their family and relationship. He or she is also more likely to like themselves, label themselves as optimists, and say that they have already achieved happiness.

Personal beliefs

• People who are bitter about Valentine’s Day also tend to be skeptics when it comes to religion, luck, or the supernatural. They’re more likely to believe that religion has brought more suffering to mankind and are more likely to say that humans are naturally evil.
• Fans of Valentine’s Day identify more as religious and attend religious services more often, are more likely to value tradition, and are more likely to say that humans are naturally good.

Overall, we see that Valentine’s Day may only hint at all the possible things one could be bitter about. On the flipside, those who love Valentine’s Day don’t necessarily need to set aside one day a year to be romantics—they do it on a regular basis already.

You don’t have to buy into cheesy romantic comedy clichés to appreciate a holiday centered on love. But it might help your life in a lot of ways to be a bit more positive about relationships, yourself, and certain days of the year. If you insist on making a statement against commercialism and “Hallmark holidays,” then maybe you should have a look at some traditions that were invented to make you buy stuff (spoiler: you’re buying more into the commercialism of Valentine’s Day by sitting at home and stuffing your face with candy hearts than by buying a card for someone). And you can always celebrate with your sweetie on the cheap or find alternative ways to have fun instead of feeding into the Valentine’s industrial complex. Remember—money can’t buy you love, but buying into the idea of love could help get you a little closer to happiness.

There’s a lot more to check out in the full report (either in HTML version or PDF version), including some surprising things about love that the Valentine’s Day lovers and bitter people have in common. The report also includes all the fun Hunch graphs for the numbers-oriented.

Dog lovers, this one’s for you.  Today we’re releasing a new report called “Like dog, like owner? What our pooches reveal about us.“  Check out the full 25 page report HERE.

The report first explores the differences – from personality to hobbies to personal beliefs – between dog owners and people who don’t own a dog. Then, it provides deeper profiles of 44 popular breeds.

But the breed profiles don’t have to do with characteristics of the breeds themselves; they instead explore the common traits shared among the people who like that particular breed.  You may be surprised about how distinct the characteristics are among the human fans of various breeds.

So whether you’re talking about members of The American Kennel Club’s more than 500 breed-specific clubs or 5,000 affiliate clubs, considering some of the impassioned canine lovers at Dogster or the readers of any number of breed-specific blogs and sites, chances are that fans of specific breeds have a lot more in common with each other that just a favorite type of pooch.

Here’s the executive summary of the report:

Compared to Hunch users who don’t own a dog, Hunch dog owners have distinct personality traits, life experiences, and demographic characteristics among dozens of dimensions:

• In general, they are more conservative, somewhat older, more likely to enjoy “mainstream” media, more likely to be female, live in the suburbs and have children, but less likely to have completed 4 years of college.
• Other differences include their greater affinity for sports and the outdoors, an aesthetic preference for rustic and “homey” furniture, and a greater likelihood of having both a security system and a firearm in their home.

Among groups of Hunch users expressing an affinity for a particular breed of dog, distinct traits again surface:

• Women tend to favor terriers, dogs with long hair, and some of the smaller breeds
• Men tend to favor the hound family and retrievers
• Other correlations are less obvious and are best explored in the main report, for example:
  • Chihuahua fans tend to be frequent doodlers
  • German Shepherd fans tend to rely more on intuition than common sense
  • Pug fans particularly enjoyed the movie The Shawshank Redemption
  • West Highland Terrier (Westie) fans think the world needs more “Family Values”, while Bull Mastiff fans would volunteer to be Tasered “just to see what it feels like” (have fun!)

There are hundreds more interesting correlations in the main analysis section of the full report.  And if your breed of choice isn’t one of the 44 profiled in the report, don’t fret.  The report’s appendix also lists all 212 purebreds and cross-breeds from Hunch’s dog breeds topic, together with links to each of their correlation reports on the Hunch site.

If this topic intrigues you, you might also check out Anahad O’Connor’s interesting NYT piece from last August.  It describes several studies exploring whether people resemble their dogs.  Fascinating reading.  And finally, for those extreme individualists out there who cringe at the thought of being grouped in a ’similarity cluster’, even if it’s innocuously based on an affinity for a particular dog breed, we have a suggestion.  For your next pooch, head on over to the nearest shelter and get a lovable mutt.  When someone asks what the breed is, you can say, “Beats me, he’s just Rover.”  But guess what?  Hunch was even able to find similarities among mutt fans. When it comes to data, you can run, but you can’t hide.

Again, you can check out the full 25 page dog breeds report in either a downloadable pdf format or the web HTML version. You can find more reports in the reports section of the Hunch site.

To date, Hunch users have added more than 40,000 tags to their profiles (if that doesn’t include you yet, step to it…).  We decided to take a look at which tags are most popular, focusing on a breakdown between users residing in North America vs. Europe.  We used a THAY question about ‘country of residence’ to get the geographic breakdown and then examined the top 200 tags by frequency for each continent.

Turns out that 13 of each continent’s top 15 tags are the same: music, liberal, food, mac, creative, books, travel, technology, funny, art, geek, movies, and photography.  In fact, of the top 200 tags on each respective side of the pond, 134, or 2 out of 3, are shared.  So, we are more ‘the same’ than ‘different’ (group hug).

But where do things differ between the two sides of the pond?

The top 25 tags used uniquely by Europeans among these ‘top 200′ data sets include: programming, london, piano, languages, jazz, italy, english, cinema, films, british, socialist, metal, it, google, clever, web design, web 20, uk, optimist, ipod, innovation, humour, german, european, and europe.  And as far as common tags that skew significantly stronger in Europe: internet, design, music, sci-fi, web, photography, philosophy, computer, atheist, rock, geek, lazy, indie, male, and football.

So I suppose you could say that Hunch Europeans are a bunch of lazy, socialist, piano-playing geeks who find themselves clever and innovative.  Or perhaps it’s more accurate to say that they’re optimistic, humourous (with an extra ‘u’) web designers who are into languages and cinema.  You pick.

The top 25 tags used uniquely by Hunchers in North America (and not appearing in the European top 200) include: artistic, family, loving, tall, married, husband, outgoing, outdoors, caring, musician, new york, nyc, hiking, athletic, libertarian, adventurous, thoughtful, teacher, american, canadian, canada, engineer, witty, spiritual, and silly.  Common tags that skew significantly stronger in North America include liberal, writer, conservative, dogs, reader, mac, father, smart, creative, green, christian, artist, cooking, travel, and video games.

So clearly, Hunch users in North America tend to be tall, liberal artists who think of themselves as caring and outgoing.  On the other hand, they’re also Christian conservatives who enjoy hiking, dogs, and the outdoors.

Probably the most accurate conclusion to all of this analysis is that you Hunchers are an incredibly diverse bunch.

Finally, we came across a nifty site called Many Eyes, developed by – of all companies – IBM.  It combines many things we love: cool data visualization techniques, data exploration, and public data sets…what’s not to love about all of that?  The graphics below are an outcome of running our tags through one of the visualization techniques on Many Eyes.

The top 200 tags among North American Hunchers:

Top 200 tags in the profiles of North American Hunch users

Top 200 tags in the profiles of European Hunch users

We had some discussion in the office yesterday about some of the posts in the Hunch forums.  We’ve been beyond pleased about the amount of forum suggestions and feedback we’ve received.  This input is invaluable in helping us define and prioritize what to keep, emphasize, refine, change, or improve.

It’s also been great to see an increasing amount of direct interaction in the forums between Hunch users.  For example, we’ve noted a growing number of forum posts from users asking for a recommendation related to a topic, with other users generally responding with good advice.  So one user might post: “Which handheld camera might be best for a low-light situation?” or “Do you think my 5 year old is ready for kindergarten even if he can’t sit still for a reasonable period of time?”  Sure, these questions could be answered by ‘playing’ Hunch’s topics in the traditional way, but – why not? – this alternative approach to posting a question and getting an answer is perfectly fine, too.

There are, however, a couple of types of questions we sometimes see in the forums which aren’t particularly well-suited to Hunch: questions of an absolute or factual nature.

An absolute question is something like: “What’s the best Chinese restaurant in Des Moines?” or “What’s the best museum in Paris?”  The first answer to questions like those is simply: “It depends.”  Are you looking for great Chinese take-out or a Chinese restaurant offering a white tablecloth experience?  Do you want a Paris museum that’s good for kids, known for contemporary art, or in a building of renowned architectural importance?

Hunch is also not the place to go to find cut-and-dry facts.  To figure out “On what day of the week does Christmas fall this year?” or “How many flight miles is it between New York and LA?”, we would cheerfully direct you to your nearest preferred search engine or answer site, which is exactly what we do for these types of queries.

So in a nutshell, Hunch isn’t meant to tell you what’s best for everyone, and isn’t meant to provide you with rote facts.  But for offering you smart recommendations that are customized to you and based on some serious math and data?  Now you’re talkin’.  And this is how we summed all that up in a phrase: “With Hunch, it’s never about simply ‘What’s Best?’, but rather: ‘What’s Best for You?’”

The Daily Beast included us in their roundup of 25 Ways to Get Smarter in 2010, just in time for 2010! Here we are, the Smarter Search Engine: Hunch.com:

A small group of MIT computer scientists has developed an algorithm called Hunch that works on a single premise: Groups of people make better decisions than individuals, even individual experts. Hunch is brilliant in its simplicity: It asks you narrower and narrower questions, and continues suggesting answers until it finds what you want. It then uses the your search to find even better answers for the next user. For instance, to the question, “What should I get my mom for her birthday?” Hunch might ask questions like, how much do you want to spend? How old is she? Is she conservative or liberal? Does she drink? In the six months since it was launched, users have already created over 5,500 decision topics and used it to purchase electronics, figure out what book to read, and plan entire vacations.

Performed by Lauren Gabriela and Peter Coles