The time, energy and money invested in a research project, and the effects of subsequent decisions, demand a very serious understanding of the underlying research process. The complete information contained in data does not always see the light of day because traditional data analysis techniques do not deal with the subtleties and complexities inherent in a research situation.

New cutting-edge techniques mean we can now address these problems instead of just having nightmares about them.

Using Item Response Theory (IRT), in particular the Rasch Model as extended by Benjamin D. Wright and his colleagues, Meaningful Measurement has developed unique methods that fulfill the most stringent scientific requirements, thereby creating additional information and new ways of looking at data.

All facets of a research situation are calibrated and explored This goes beyond the conventional investigation of questions and respondents to include assessment using raters to judge various situations or products.

The following is a brief description of Meaningful Measurement’s preferred methodology.

Meaningful Measurement is dedicated to helping organizations make the best possible decisions. In order to do so, we go beyond a traditional statistical analysis and use objective measurement techniques. This type of data analysis produces the facts of measurement, thus allowing a deeper understanding of the structure of the research question.

Scientists design, build, and calibrate instruments to record physical phenomena. When latent trait variables such as “Attitude” or "Ability" are measured indirectly, fundamental objective measures must be constructed with which to measure the underlying dimension.

Unfortunately it is not possible to grab a piece of attitude or ability and measure it with a ruler. Therefore psychometricians must take great care to construct a frame of reference which evokes these objective, standardized measures. Only then can data be interpreted.

Objective Measurement requires the following:

• An underlying trait that can be expressed in terms of more or less

• Survey/test items are the operational definition of the underlying trait

• Survey/test items can be ordered from easy to hard

• Respondents can be ordered from less to more in attitude or ability

Rasch/IRT Model:
Meaningful Measurement uses the techniques of Item Response Theory (IRT), in particular the Rasch model one parameter logistical model (1PL) which meets the requirements for measurement. This method is widely used in educational testing, certification and licensure, outcomes assessment, and many other research applications.

P1,0 =          e (ability-item_difficulty)
                    1 + e (ability-item_difficulty)

Advantages of Using Item Response Theory:

• Equal Interval Measure

• Test/survey-takers and items are represented on the same scale

• Item calibrations are independent of the respondents used for calibration

• Respondent ability/attitude estimates are independent of the particular set of items used for estimation

• Measurement precision is estimated for each person and each item

Rasch has the added value of diagnostic methods to investigate an emerging field. Direct comparisons of communities are possible with the standard metrics. This precision allows Subject Matter Experts to make the most informed decisions possible.

Data Analysis
The computer programs Winsteps 3.52 or Facets written by John Michael Linacre, provides the basis for data analysis. Once raw scores are conditioned into measures, traditional statistical analyses are performed. Additional analyses, charts and graphs are produced by SPSS 18.0, Excel, and PowerPoint.