How does Classworks Calculate Rate of Improvement (RoI) for students?

How does Classworks Calculate Rate of Improvement (RoI) for students?

Classworks takes the median of a student's first three data points as a starting place for the student. Using the least squares regression model, the student's rate of improvement is calculated. 

What is the least squares regression model? 

The least squares regression model is a formula that helps us understand and quantify a student’s rate of improvement over time. It’s like drawing a trend line through a series of data points (like progress monitoring scores) to see how much a student is progressing.

Key Components:

Trend Line (Best Fit Line):

  1. Imagine you have a graph with time (weeks, months) on the x-axis and student performance (scores, number of problems solved) on the y-axis.
  1. The trend line represents the general direction of the student’s performance over time.

Slope:

  1. The slope of the trend line shows the rate of improvement.
  1. If the slope is positive, it means the student is improving. The steeper the slope, the faster the improvement.

Intercept:

  1. This is where the trend line starts on the y-axis. It represents the student’s starting performance level before the improvement trend begins. With Classworks, it is the median of the first three test scores. 

How It Works:

  1. The model takes all the performance data points and finds the line that best represents the overall trend.
  1. It minimizes the differences between the actual data points and the trend line, ensuring the line is as close to all the points as possible.

The Formula

  1. The formula for the least squares regression model, specifically for simple linear regression, is: y=mx+b 
  1. where:
    1. y is the dependent variable (the outcome or response variable).
    1. x is the independent variable (the predictor or explanatory variable).
    1. m (also denoted as β1) is the slope of the regression line, which represents the change in y for a one-unit change in x.
    1. b (also denoted as β0) is the y-intercept, which represents the value of y when x is zero.
  1. To find the values of m and b, we use the following formulas:
  1. m=n∑(x2)−(∑(x))2n∑(xy)−∑(x)∑(y)
  1. b=n∑(y)−m∑(x)
  1. where:
    1. n is the number of data points.
    1. ∑(x) is the sum of the x-values.
    1. ∑(y) is the sum of the y-values.
    1. ∑(xy) is the sum of the product of each x-value and its corresponding y-value.
    1. ∑(x2) is the sum of the squares of the x-values.
  1. These formulas are derived by minimizing the sum of the squared differences between the observed values and the values predicted by the linear model, which is the essence of the least squares method.
Notes
Least squares regression model is one of the most widely accepted methods of calculating student ROI for progress monitoring.