Comparing the DOT, O*NET and eDOT

#3: What Measures Are Used?

For occupational classification data to be found reliable, you must also be able to defend its measures in court.

O*NET's measures

O*NET analyses include the collection of several hundred descriptive ratings based on questionnaire responses. Over 277 primary descriptors include traditional measures such as “Standing” and “Sitting.” In addition, new variables such as “Competitive” and “Conflict Situations” refer to cognitive and psychosocial demands of work that are new to job analysis.

Because O*NET measures are so numerous, we refer the reader to the O*NET Internet site for their review (O*NET Data Collection Program).

Vague measures. Many of the O*NET descriptors are not straightforward. Some descriptors are difficult to observe in the work place and difficult to relate to the claimant or prospective worker.

Skipped measures. If an incumbent reports the requirement is "not important," then O*NET skips obtaining measures for that job requirement.

Example: An incumbent fills out an O*NET questionnaire. He rates a job requirement as not important, even though an albeit low level of the requirement is needed to perform the work.

This reporting of low-level requirements becomes especially crucial when occupational data is used for disability analyses. Modest levels of a requirement (such as stooping) may preclude an individual's employment, even when the requirement does not seem important to a job incumbent.

Irrelevant measures. Incumbent respondents could easily report skills and abilities that are not necessary for the occupation. This makes the use of O*NET data problematic.

Validity not determined. Finally, concern exists regarding the descriptors validity:

Do they accurately reflect the type and level of functioning required in an occupation?

When the Social Security Administration evaluated O*NET for use in disability determination, it found that it was developed for career development and exploration purposes and is not suitable for disability evaluation.