Incremental Testing

What is incremental testing?

In the software world, incremental testing is a way of integration testing in which you first test each module of the software individually, then continue testing by adding another module to it, then another. The advantage is that when you encounter an error you can find the defect more easily because you have a better idea in which subsystem to look. We can take this method and apply it to testing in our educational system.

Incremental Testing

  • is technology driven
  • can be assessed more often
  • measures personal improvement
  • is continued from year to year
  • boosts student confidence
  • reduces teacher stress
  • eliminates “test teaching”

The Basic Concept

The test questions for each subject are separated into varying degrees of difficulty. The education commission, principal, teacher, whomever, chooses some number of degrees to cover in each subject matter, each education level and how many correct answers it takes to end each test section. The test questions begin at the lowest difficulty level (for that subject and that particular education level) and increase or decrease incrementally depending on the answer. For every correct answer, the next question will be more difficult, same degree. For every two correct answers, the next question will be the next higher degree of difficulty, and vice versa. For each incorrect answer, the next question will be easier, same degree. For every two incorrect answers the next question will be the next lower degree of difficulty. The student jumps from degree to degree until he/she obtains the required number of correct answers.

Example

Our teacher, Linda, has chosen to cover 3 degrees of math this year. Linda logs into the app and brackets the degrees she has chosen to cover. For this particular math class she selects 3, 4, and 5. She decides each student must get 10 correct answers to finish the test. Once she has finished all her selections, Linda generates the test. On test day, our students arrive and log into the application with their unique credentials. Like any other test they get pencil and paper to work out problems. One student, Mark is especially nervous. He has had trouble with math tests in the past. He is worried he will not pass this new test. The test begins with questions at degree 3. Mark answers two questions correctly which moves him to degree 4. He answers two more questions correctly which moves him to degree 5. He answers one question correctly and then misses two questions which moves him back down to degree 4. Mark gets one right, one wrong, one right, two wrong, which moves him to degree 3 where he answers two more questions correctly. This moves him back up to degree 4 for another correct answer which ends the test. It has taken Mark fifteen questions to get ten correct answers. Linda receives a report which details how many correct and incorrect answers Mark submitted for each degree. She can even view the exact questions he missed. It is apparent that Mark needs some more work on the 5th degree subject matter. Mark receives a report which details the percentage of correct answers he submitted for each degree. Mark is relieved that there was no score, but he also see that he needs more work on 5th degree subject matter. Mark is determined that next time he will get more right answers on the harder questions.

I have diagrammed three students’ test scores according to the incremental testing method. All students ended with the required 10 correct answers. The total number of questions it took to finish the test were different for each student, however, the difference is negligible (2 or less). As my chart indicates, student 1 made average progress, remaining in the middle; student 3 made great progress, even answering questions above our testing bracket; while student 2, on the other hand, is slightly below average and may need some more help.

Why use incremental testing?

Incremental testing is rooted in technology. It is portable, reusable, and upgradeable all with the click of a button. It saves money, space, time and trees by occupying virtual space. No thousands upon thousands of booklets and test sheets to hand out. Testing apps can even be installed on student’s own laptops, tablets and smartphones. Then, when it is time to take the test, the facilitator unlocks it. Digitally aided testing helps reduce cheating. By using a database of questions sorted by subject and difficulty degrees, each student gets questions tailored to their specific ability levels. Even if two students are on the same level they may have different questions.

Because incremental testing is technology based, teachers and schools can conduct tests more frequently. Students can be evaluated from school year to year. Or at the beginning, middle, and end of each school year. Or even once a month. Evaluating students more frequently helps educators understand their students’ knowledge retention and personal growth better than traditional pass/fail testing methods. In the real world we are rarely, if ever, in pass/fail situations. We do not get grades or scores. We either get our job done or we don’t. Some of the most useful real-world skills like critical thinking and problem solving aren’t even measured on standardized testing.

Incremental testing helps boost student confidence. Instead of worry about passing, students focus on improvement. They see how well they did in the past, and work towards moving up the scale. Just like in many games, the goal is to get to the hardest level possible. Let’s face it, not every student is great or wants to be great at everything. Some are better at English. Some are better at math. Some students could care less. Some students are smart but terrible test takers. With incremental testing there are no good or bad, smart or dumb students. There are strengths and weaknesses. Mark doesn’t have to feel dumb because he got 7 questions wrong and Jenny had none. Instead Mark realizes he is better at algebra than Jenny, but Jenny has a better grasp of geometry.

Incremental testing eliminates “teaching to the test.” Teachers do not have to worry about whether or not every student makes a 100% on the test. They do not have to worry about rushing through a boatload of material that has to be covered because some committee dictated that is what kids of this age should know. Everyone learns differently. Standardizing our citizens is no way to advance our population’s intelligence or abilities. By measuring a student’s and teacher’s performance based on percentage improvement, rather than a set score, teachers can relax and focus on helping their students no matter where they are on the learning curve. We should be celebrating student’s individual talents, not trying to stuff them all into the same mold.