Students as Researchers

Students are not only learners in schools they can be researchers about their learning progress. Listed below are ways students have evidence of learning or evidence of not learning. There is no particular order for these conversations with the student researchers.

  1. Look for class run chart patterns. Let the students know that data needs patterns in order for scientists to draw conclusions. Thus, one or two weeks of improvement is nice, but is not a trend. Likewise, one or two weeks of a lack of improvement is not a trend. When the class run chart is moving in the same direction for 3 or more quizzes, the basis for deep classroom conversations is possible. The reason for going up or down three times in a row is unlikely to be bad or good luck; something else is occurring.
  2. Be sure to request the free LtoJ Effect Size Calculator by writing You will receive a link to a google.doc. The updated scatter diagrams on the LtoJ website shade the columns for data that is used for the effect size calculation. Remember, students enter the data into the calculator. One reads the numerals off the scatter diagram and another enters the numerals into the calculator. A third student is often involved is double checking accuracy. At the top of the calculator is a chart helping students compare their learning to the average learning from John Hattie’s 250 meta analyses. The photo of Angela Willnerd on this page shows her helping her grade 1 “student researchers” understand effect size. The student researchers want to know what they are doing that is creating either the wonderful or the disappointing effect sizes.
  3. Usually the graph for item analysis is started second quarter with students graphing errors from most missed to fewest missed. The ideal discussion for these researchers is to delve into why a review question was missed by so many students.
  4. Student researchers get to establish a hypothesis for their improved learning. When the class run chart is flatlining, conduct a class meeting. The students generate ideas for getting the class run chart back on a growth trajectory. These ideas should be for classroom implementation instead of work at home. After a list is compiled, the students vote on the hypothesis they most want to test. Then the teacher allocates time for the students to carry out their experiment. Typically 10 minutes a day, 4 days a week, for 3 weeks is allowed for the research. At the end of the three weeks, if the class run chart has not improved, have another class meeting to determine a different hypothesis to test. If the class run chart shows improvement, keep the experiment going until it loses its effectiveness.
  5. Metacognition is a good word. However, for the purposes of these activities it is better to refer to these activities as student researchers.

Above is a picture of Angela Willnerd, grade 1 teacher in Fremont, Nebraska. She is explaining effect size to her students at the end of third quarter. In her left hand are 4 unifix cubes. They represent the 0.40 average effect size from John Hattie’s research of 250 influences upon learning. In her right hand are 19 cubes. These represent an effect size of 1.9 at the end of third quarter in spelling. Her students are performing at almost 5 times the average with a full quarter of school left. Angela utilizes LtoJ® for her spelling program. There is no weekly quiz to cram; the Friday assessment is 12 words randomly selected from the 150 for the year. Thus the 1.9 represents words in long-term memory. You can watch her spelling in action here.