Wednesday, May 25, 2016

Right Mark Scoring

Everyone knows traditional multiple-choice or right mark scoring (RMS): students mark, teacher scores, and that is it. Stage One is done. 

In Stage Two the teacher may review difficult questions. Then, if an item is thoroughly discussed and of great value give all students the point. If it turns out to be just a bad item, drop it. Click Score for the "true" test and scores embedded in the full test. You have made a lower level of thinking STANDARDIZED classroom test. Stage Two is done.

In Stage Three, a genuine learning environment is established by students and teachers discussing items classified as Expected, Discriminating, Guessing, and Misconceptions. This is only possible with Knowledge and Judgment Scoring (KJS). Stage Three ends with re-scoring to produce a higher level of thinking STANDARDIZED classroom test.

The test fitness is an estimate of how well the test fits student preparation. This is the average score if all students reject known wrong options and then guess from the remaining items on each item. In operation, multiple-choice tests are easier than their design value, such as, one out of four, 25%. Test wise students know this.

The omit value can still be adjusted in Power Up Plus, built on Break Out and advanced features. The Omit default value for Knowledge and Judgment Scoring is set to 50% in Power Up Plus.

[Copyright dates are all 2006 when the company was created. In 2013 the company copyrights were returned to me.]

Wednesday, May 18, 2016

The Company - 2006-2013

My experience with most faculty members who use computer scoring is to find the sheets containing the scores and toss the rest of the printout into a waste basket. I therefore added teacher and student friendly printouts to Break Out to market Knowledge and Judgment Scoring. These did not require interpretation or an understanding of educational statistics.

Work on hosting Power Up Plus as an interactive scoring and analysis service stopped when it was realized that no one person could create and maintain the needed NCLB package of item banking, scoring, analysis, and item ranking; even with the minuscule results obtained from traditional right mark scoring, let alone all the additional results available for student, teacher, instruction and test development with knowledge and judgment scoring. 

No child left behind (NCLB) promoted administrative direction of what, when, and how instruction would take place. "Every class on the same page, on the same day, by a prepared script." regardless of student preparation and ability to learn. It had strong political appeal. Common Core State Standards (CCSS) attempts to restore teacher-student interaction and student development, but then falls again into the gutter of best intentions by misusing multiple-choice tests such that they defeat what CCSS attempts to do: to increase levels of thinking, not to continue to appeal to a lucky test day. Again politics and money.

I left several research features in the program including the selection of the portion of the class test score that comes from quantity and from quality. These may be of interest to advanced students, teachers, and administrators wanting a new area of research: updating traditional multiple-choice scoring so that testing guides student development rather than hinders it. These may be of use again now with the demise of NCLB and CCSS.

Wednesday, May 11, 2016

Break Out - Free Software

A flood damaged a fence on my parent’s farm. I helped my dad clear and rebuild it. Poison ivy was everywhere. No need to worry, as I had played in the stuff in elementary school.  A just finished two year leave in Hawaii with the Agriculture Research Service (1966 - 1968) had changed things. What followed was a systemic reaction.

One recommendation was to keep cool and not sweat. I knew where that could be found: the university library (65 degrees in summer and 85 degrees in winter) and with two new computer terminals. With an instruction manual in hand I learned to program the campus computer.

It was in the fall of 1981 that I first wrote my own scoring and analysis software. I had just returned from a two-year leave with the Environmental Protection Agency in Colorado. I learned what computers could do for the testing of pesticide applicators.

This started the Net Yield Scoring software that became Break Out: Break out of the habit of forced-choice marking and counting right marks. Break Out has always been free software (Dropbox/Nine-Patch Files/download/BrkOutSC.ZIP).

Wednesday, May 4, 2016

Printouts - Index

Traditional multiple-choice, right mark scoring (RMS) is generally carried out in a static mode. Students are a product being assessed by the number of right marks. The score is a lower level of thinking performance rank on that test.

Knowledge and Judgment Scoring (KNS) allows student an equal voice in testing with the test makers and teachers. There is a big difference between testing those who have learned and those who are learning. Master teachers know this and respond to student needs.

KJS is dynamic. Learning is dynamic. Knowing is static. Rote memorizing what some one else knows is static. 

Questioning is dynamic. Proposing answers is dynamic. Verifying  answers is dynamic. Having learned is static. A different right answer may exist at each stage to the same question. This is why it is so difficult for those who know to guess how difficult a question my be on any one test (RMS or KJS). Knowing requires an analysis beyond RMS. 

Knowledge and Judgment Scoring can assist every teacher to be a master teacher: to know what each student knows he/she knows and what has yet to be learned. It then follows that high quality students get high test scores. They can answer questions they have not seen before.

The following printouts illustrate RMSing in a small class and KJSing in a large class.