Last Saturday evening, I hosted a dinner party. I wanted it to be the perfect dinner party. Long before I sent out the invitations, I fastidiously planned for a three-course meal.

While many budding young chefs turn to the Internet for recipes, I sought out a more authoritative source. During the summer, I discovered the TV program “America’s Test Kitchen” on PBS. Christopher Kimball, editor of Cook’s Illustrated magazine, and his team of chefs approach the culinary arts like a science. They experiment with different recipes and conduct blind taste tests for every recipe they publish. The show even has an in-house chemist!

“Our recipes are perfect and foolproof because they have been tested again and again,” Kimball often proclaims.

“America’s Test Kitchen” appealed to the inner nerd in me because, at the time, I was working for Columbia political scientist Donald Green. Green and his colleague, Alan Gerber, a professor at Yale, pioneered the use of experimental methods in their discipline. For instance, to test the effectiveness of “get-out-the-vote” campaign strategies, they conducted trials in dozens of American cities. Through these field experiments, Green and Gerber discovered what campaign tactics work (canvassing, nonpartisan mail and phone banks) and what tactics didn’t work (leaflets, partisan mail, robocalls and emails).

If experimental methods could improve political campaigns, then surely they could make my dinner a huge success. All I would have to do is follow the directions in the recipe book written by Kimball and his team of mad scientists.

Except I got to the Calhoun College kitchen an hour too late.

Except the kitchen ran out of black pepper.

Except I could not find measuring spoons.

Forty minutes into cooking, the kitchen was a mess. Plates of chicken juice sat on the counter. The oil in the frying pan popped in mini-explosions. Because I forgot to poke holes in the potatoes before I put them in the oven, I anticipated they were going to turn into grenades.

At last, the dinner was completed. I sighed in disappointment (also in relief because the potatoes didn’t explode). I’d say the meal averaged about a “Yale B+.” The chicken, though incredibly crispy, sat in a little pool of grease. The carrots, though buttery, lacked flavor. The peas, though well-seasoned, turned cold and mushy. My entree paled in comparison to the immaculately decorated chocolate cake my friend brought for dessert. My replication of a culinary experiment failed.

Despite dozens of reassurances about my cooking, I remained sullen. What had gone wrong?

Looking at my friends’ stuffed mouths and distended bellies, I realized maybe nothing had gone wrong after all. In a double-taste test, the subjects would have undoubtedly pooh-poohed my meal. But I was among friends who needed a respite from the dining hall and Gourmet Heaven. As much as I would like to believe that cooking is a science, I have to admit it is partly an art. It is as much about creating and enjoying food with the people you love as it is about weighing flour or taking the temperature of meats.

I hate to turn into Rachael Ray or Paula Deen, who spin platitudes about life using culinary metaphors, but their folksy wisdom (and inexact measurements) have a point. We live in an era of big data where a lot of our personal lives can be quantified: how many friends we have on Facebook, how many photos we are tagged in, how many people “liked” our status updates or who is likely to vote in a presidential election. We have become SAT scores, GPAs or applicant #498 for a job. Our human interactions are seemingly mapped out with precision.

English poet Alexander Pope lived during the Enlightenment, an era of important scientific and technological discoveries. In his poem “The Proper Study of Mankind,” he praised science yet remained skeptical about its ability to explain humanity. One can “Go, measure earth, weigh air, and state the tides,” he mused, but humans are “the glory, jest and riddle of the world.”

As much as I cringe at measurement error and value experimental methods, I still like to believe the proper study of mankind is man.

Contact Baobao Zhang at .