Real-Life Decision Making -- Solution
You adjust the data and write a report supporting the hypothesis.
You think to yourself, "I am going to reinterpret the data. I am completely
certain that the hypothesis is true and that something went wrong with the
data collection. If I don't support the hypothesis, the university will be
disappointed, the sponsors will be disappointed and I will never be able to
publish the results."
The university and your sponsors are pleased. You quickly find a publisher
for your work.
However, before long, other researchers from different universities try
to duplicate your findings using the same methods and procedures. Each one
of them announces that they cannot duplicate your findings, and that their
research indicates that the hypothesis is not true.
Your university is embarrassed and your sponsors reject your application
to fund another project. Although no one accuses you of fudging data, you
know that they have lost faith in your abilities. It is harder for you to
get research projects approved in the future.
When the data doesn't fit your hypothesis the next step is clear, and it
never involves fudging the data: "Figure out the hypothesis that does fit,"
says archeology professor Doug Bamforth. "If you look at the world and it
doesn't tell you what you're expecting to see, it means you learned something
new."