Data analysis

As a researcher you analyze and interpret data.

Data analysis

You need to be critical. It is important to recognize  limitations and whether the retrieved data is biased.

The research question determines study, recruitment and reporting.

It is important to determine whether the research data you retrieve is biased.

Randomization allows you to draw a number of conclusions. However, this does not mean that a randomized study always provides correct evidence to the research question.

Blinded research is often regarded as superior. Is this the case?

Despite the fact that a randomized controlled trial (RCT), by definition, can’t exist without one, the control group is the forgotten stepchild of the RCT.

Measurement is fundamental to science. However, there is more to measurement than meets the eye.

As there are often numerous measures available for any given construct, how do you choose which to use?

Scientific writing is by nature technical, so getting terminology and wording right is important. When there is inconsistency in definitions, or misuse of words, problems follow.

Scores on outcome measures matter in the clinical world because they influence treatment decisions, and because payers are increasingly asking clinicians to justify their treatment decisions.

More and more software tools for data analysis are becoming available. You can download tools via the HU Software Center.

To judge whether one treatment is more effective than another, simply knowing whether a difference exists is not enough. We need to know how big the difference is.

Confidence intervals: linking evidence to practice.

Software tools for data analysis

Do you have questions on analyzing qualitative data ATLAS Ti?  Please contact Marlies Welbie.

Switch to English English
× Text the desk