How Do People Decide – The Order Effect

I like to think that when I decide to buy something it’s because I’ve thought it through and it’s the best purchase. Logically and rationally. But I’ve read too much research on the unconscious and on decision making. So I know that people, including me, make decisions based on unconscious processing and that I don’t use logic as much as I’d like to think I do.

There’s this great research study that proves two concepts: 1) We don’t pay attention to very many attributes about a product, 2) We tend to choose the first item that appears on a product page of a web site.

Here’s the research citation:
Felfernig, A., G. Friedrich, B. Gula, M. Hitz, T. Kruggel, G. Leitner, R. Melcher, D. Riepan, S. Strauss, E. Teppan, and O. Vitouch. 2007. Persuasive recommendation: Serial position effects in knowledge-based recommender systems. In Persuasive Technology, Second International Conference on Persuasive Technology. New York: Springer.

Felfernig set up a website with tents. He had visitors to the site fill out a questionnaire about the type of camping they planned to do. Then the site recommended four tents based on ten different attributes, such as waterproofing, weight, air ventilation, etc. Based on the questionnaire you filled out, two of the four tents are rated as “best buys” for the attributes that are important to you. Which tent will you buy? Tent 1, 2, 3, or 4?

Even though there were ten attributes that the tents were compared on, participants in the study paid attention to only 2 or 3 of the attributes. And when it came time to pick a tent, the participants in the study didn’t even consider the attributes. They picked the first tent more than 2.5 times than any other. They chose the first one 200 times versus 60 for all the other choices combined.

But, just like me, the participants explained their choice, based on the logical decision they thought they were making. They would explain the choice of tent #1 by saying, “This tent is the most waterproof”, for example. They thought they were weighing all the attributes of all the tents, but in reality they were only looking at a few attributes and even those didn’t matter… all that mattered was which tent showed up first.

I’m going to see if I can pay attention to this the next time I go to buy something at a website where there are several choices on the page at once. Maybe if I notice I’m doing it I won’t be so prone to doing it?

top10cta

New York City Stabbing a Hoax?

I was just doing some research on a murder that happened in Queens NY in 1964. You may have heard of it. It’s the Kitty Genovese murder. It’s the crime that led to an entire branch of social psychology research.
Kitty Genovese was stabbed to death on the street while 38 witnesses watched and did nothing to help. Social scientists became fascinated by what they called the “Bystander Effect” and a whole series of research studies began to study why it is that people will take action to help when they are by themselves, but not if they are part of a group.
I’m writing about this in a book I’m working on (Neuro Web Design: What Makes Them Click, PeachPit, due out in Jan of 2009). I found online the original New York Times article about the murder, but then I found online another New York Times article written 40 years later in 2004 that casts doubt on some of the data and its interpretation of the original event. 
Apparently it’s now believed that several people probably heard something and maybe saw something, but they probably couldn’t have figured out what they were hearing or seeing (based on where the crime occurred and the lighting on the street etc), and it probably wasn’t 38 people either. So the truth is that a few people heard some noises and saw someone staggering down a street.
The question I have is: If it took me about 5 minutes to find this updated information on the internet, then why does the original version of events still show up? In research articles, in slide presentations, in books, people still talk about the Kitty Genovese event without mentioning the later update. 
Is it the sheer number of references to the earlier, incorrect version? Or is it that everyone is lazy and they take the first reference they come across?