Empirical Investigation

There are many ways to empirically investigate

But it’s generally better to do it sooner than late

Even consider having a plan

Than rave around with an empty tin can

Put down a hypothesis

It might come to you when you’re taking a piss

A hypothesis is a statement, a bit like a promise

That you aim to keep and hope not to miss

Now Jonathan, Jinjuan and Harry have written in their testament

It is a “focused statement that can be examined by a single experiment”

And single means one and not a several or a few

So be sure the hypothesis isn’t some muck you pulled of your shoe

There are several research methods from which to choose

Choose wisely, or you might loose clues

I read in their book and here’s (more or less) how it plays:

Investigation can generally be described in three different ways

I’ll describe it here and now, hence forthwith

You can conduct a descriptive investigation with

Observations, Surveys and Focus Groups

But be prepared before you call the troops

Or try out relational investigations

To identify connections between multiple relations

But it won’t help you discover the causal effect

No relations are ever easy or perfect

Experimental investigations help to find out the latter

Where you conduct experiments to uncover the matter

You can begin in a laboratory or take it to the street

Do you it on your friends or random people you meet

There are many research methods available to you

But as usual, it is up to you to know what to do:

You can try out observations, surveys and focus groups

Or field studies, interviews and usability studies

And to find out the causal effect

Try controlled experiments to see what you get.



Jonathan Lazar, Jinjuan Heidi Feng and Harry Hochheiser (2010) Research Methods in Human-Computer Interactions, 20-40.