View Meta-Analysis Database

1District
1School
1Teacher
2Grade
2Level
3Subject
3Topic
4Strategy ID
4Target Strategy
5Unit Length (Days)
6N (Control)
6N (Treatment)
7Probability
8Effect Size


Column descriptions:

1 The columns labeled District, School, and Teacher contain numeric values representing each district, school, and teacher listed in the data file. These values are entered sequentially. For example, as more action research studies are added to the data file, each teacher will be assigned the next number in the sequence.

2 The column labeled Grade contains the grade or grade band as reported by the teacher who participated in the study. The column labeled Level contains the following values: ES – Elementary School, MS – Middle School, or HS – High School.

3 The columns labeled Subject and Topic contain the general subject area and unit topic as reported by the teacher who participated in the study.

4 The column labeled Strategy ID contains a numeric value representing the target instructional strategies present in the data file. The column labeled Target Strategy contains the target instructional strategy as reported by the teacher who participated in the study. Based on the information in this column, a value is assigned for the Strategy ID column. When a teacher reports using more than one target instructional strategy, the teacher’s data is reported in the data file with each Strategy ID.

5 The column labeled Unit Length (Days) contains the length of each unit in days as reported by the teacher who participated in the study.

6 The columns labeled N (Control) and N (Treatment) contain the number of students in the control group (target instructional strategy NOT used in the classroom) and the number of students in the treatment group (target instructional strategy used in the classroom), respectively.

7The column labeled Probability contains the p-value for each study in the data file. Basically, if the value in this column is less than .05, the effect size reported for the study can be considered statistically significant at a significance level of 5% (α = .05). In other words, a reasonable inference can be made that the reported effect size is probably not a function of random factors; rather, the reported effect size represents a real change in student learning. For a more thorough discussion of significance testing, see Harlow, L.L., Mulaik, A.A., & Steiger, J.H. (Eds.), (1997).What if there were no significance tests? Mahwah, NJ: Erlbaum.

8The column labeled Effect Size contains the calculated effect size for each study in the data file. The calculated effect size is the standardized mean difference between treatment and control groups. Stated differently, it is the difference in the average score of the treatment group and the control group stated in standard deviation units. Unless noted otherwise, the effect size for each study was computed as Cohen’s δ using the following formula: