Information Brief
Evaluating Student Performance Using a Trend Line Analysis: The Tukey Method
During the progress monitoring process, educators analyze students’ data to evaluate their performance and adjust instruction as needed. One way they can do this is by using a trend line analysis. When using this method, educators compare the following for an individual student:

Goal line—the straight line representing a student’s expected rate of improvement needed to meet a predetermined goal
Trend line—the straight line representing the general direction of the student’s scores
Though most software programs will graph a trend line, educators can use the Tukey method to graph a trend line by finding the line of best fit for a student’s data across a given period. While it may appear complicated at first glance, the Tukey method is quite simple, requiring only graph paper and a pencil.
Step-by-Step Instructions
For Your Information
A minimum of eight data points is needed to conduct a trend line analysis. For example, to gather this data, educators can collect data:
- Once per week across eight weeks of instruction
- Twice per week across four weeks of instruction
Following are step-by-step instructions for using the Tukey method to create a trend line. Although the Tukey method can be used in all subject areas, for illustrative purposes, these instructions incorporate a student’s reading scores. The table below contains the student’s scores for eight weeks of instruction (plotted each Tuesday), and the graph visually represents the student’s performance across the eight weeks in relation to the goal line.
| Weeks of Instruction | Words Correct |
| 1 | 9 |
| 2 | 12 |
| 3 | 14 |
| 4 | 11 |
| 5 | 15 |
| 6 | 14 |
| 7 | 12 |
| 8 | 15 |

Step 1

Divide scores into three relatively equal groups.
If possible, divide the scores into three equal groups. If the scores cannot be divided equally, divide them into approximate thirds.
In this example, there are eight scores (or data points), so they cannot be divided equally. To divide these scores into approximate thirds, draw two vertical lines—denoted in pink—that place three scores in the first group (9, 12, and 14), two in the second group (11 and 15), and three in the third group (14, 12, and 15).
Step 2

Find the intersection of the midrate and middate for the first group of scores.
For the first group of scores, determine the midrate or median score—the score that falls in the middle when scores are arranged from lowest to highest. Next, identify the middate or the middle week of instruction. Then, place an X at the intersection of the midrate and the middate.
In this example, there are three scores across the first three weeks of instruction. The scores are already arranged from lowest to highest, so the median score (midrate) and the middle week of instruction (middate) are easily identified.
Median score (midrate): 9, 12, 14,
Middle week of instruction (middate): Week 1, Week 2, Week 3
Now, place an X at the intersection of the midrate (12) and the middate (Week 2).
Step 3

Find the intersection of the midrate and middate for the third group of scores.
For the third group of scores, determine the midrate and middate. Then, place an X at the intersection of the midrate and the middate.
In this example, there are three scores across the final three weeks of instruction. Once these scores are arranged from lowest to highest, the midrate and middate can be easily identified.
Median score (midrate): 12, 14, 15
Middle week of instruction (middate): Week 6, Week 7, Week 8
Now, place an X at the intersection of the midrate (14) and the middate (Week 7).
Step 4

Draw the trend line.
Draw a line beginning at the X in the first group and continuing through the X in the third group.
In this example, a line is drawn from the first point (the intersection of 12 and Week 2) to the second point (the intersection of 14 and Week 7) to create the trend line.
Note: Because the Tukey method only uses the median scores of the first and third groups, it does not take into account all the data points and is, therefore, an estimation of the trend line.
Making Data-Based Instructional Decisions
After using the Tukey method to graph a trend line, the educator can use a trend line analysis to evaluate student performance. Determining whether the student’s trend line is steeper, flatter, or around the goal line enables an educator to make a data-based instructional decision, sometimes referred to as data-driven instruction (DDI). The table below illustrates how to use a trend line analysis to interpret student data and, in turn, make an informed instructional response.
| Position of the Trend Line | Instructional Response |
Steeper than the goal line…![]() |
Increase the goal—If the student’s trend line is steeper than the goal line, the student’s performance is exceeding expectations and a more ambitious goal is needed. |
Flatter than the goal line… ![]() |
Revise instruction—If the student’s trend line is flatter than the goal line, the student is not making adequate progress. The educator should implement a different instructional approach and continue to monitor the student’s progress to determine if the approach is effective. |
Around the goal line… ![]() |
Make no changes—If the student’s trend line and goal line are about the same, the student is on target to meet the goal. No changes to instruction are necessary at this time, but the educator should continue collecting data. |
Video Example
In the video below, the educator follows the step-by-step instructions described above to accurately and efficiently draw a trend line. After doing so, the educator compares the trend line and goal line to evaluate the student’s performance.
Transcript: The Tukey Method
Evaluating Student Performance Using a Trend Line Analysis: The Tukey Method
Educators can use trend line analysis to evaluate student performance and make data-based instructional decisions. When using this method, the educator compares the student’s goal line and trend line. The goal line is the straight dotted line representing a student’s expected rate of improvement needed to meet a predetermined goal. It may be denoted by a capital G. The trend line is the straight solid line representing the general direction of the student’s scores. It may be denoted by a capital T.
Before conducting a trend line analysis, the educator should collect at least eight data points. For example, educators can collect data once per week across eight weeks of instruction or twice per week across four weeks of instruction. Once these data are collected, the educator can easily and quickly create a trend line using the Tukey method. Let’s look at an example.
An educator collected data once per week for eight weeks and graphed the following scores: 9, 12, 14, 11, 15, 14, 12, and 15. The educator then used the following steps to create a trend line.
Step 1: Divide scores into three relatively equal groups. If the scores cannot be divided equally, divide them into approximate thirds. In our example we have eight scores, so we divided them into approximate thirds. We do this by drawing two vertical lines—denoted in pink—that place three scores in the first group, two in the second group, and three in the third group.
Step 2: Find the intersection of the midrate and middate for the first group of scores. The midrate or median score is the score that falls in the middle when scores are arranged from lowest to highest. The middate is the middle week of instruction. Place an X at the intersection of the midrate and middate. In our example, the first group contains three scores across the first three weeks of instruction. The scores are already arranged from lowest to highest, so we can easily identify the midrate and the middate. Now we place an X at the intersection of the midrate (12) and the middate (Week 2).
Step 3: Find the intersection of the midrate and middate for the third group of scores. In our example, the third group consists of three scores across the final three weeks of instruction. Once we arrange these scores from lowest to highest, we can easily identify the midrate and middate. Now we place an X at the intersection of the midrate (14) and the middate (Week 7).
Step 4: Draw the trend line. Draw a line beginning at the X in the first group and continuing through the X in the third group. In our example, we draw a line from the first point to the second point to create the trend line. We then label it with a capital T.
After using the Tukey method to graph a trend line, you can use a trend line analysis to evaluate student performance. If the trend line is steeper than the goal line, the student’s performance is exceeding expectation and a more ambitious goal is needed. If the trend line is flatter than the goal line, the student is not making adequate progress. As such, the educator should implement a different instructional approach and monitor the student’s progress to determine if the approach is effective. Finally, if the trend line is around the goal line, the student is on target to meet the goal. No changes to instruction are necessary at this time, and the educator should continue collecting data.
Looking back at our example, we see that the student’s trend line is flatter than the goal line. The educator will implement a different instructional approach and monitor the student’s progress to determine if this approach is effective.
Fuchs, L. S., & Fuchs, D. (2011). Using CBM for progress monitoring in reading. National Center on Student Progress Monitoring. https://files.eric.ed.gov/fulltext/ED519252.pdf
National Center on Response to Intervention. (2011). Module 2: Progress monitoring training manual. Beginning implementer series. https://files.nwesd.org/jlongchamps/RTI/Module%20 Two/PM_Manual-ed_clean%209%20211%20(2).pdf
National Center on Response to Intervention. (2013). Brief #3: Common progress monitoring graph omissions: Making instructional decisions. Progress monitoring brief series. https://files. eric.ed.gov/fulltext/ED578046.pdf
Parker, R., Tindal, G., & Stein, S. (1992). Estimating trend in progress monitoring data: A comparison of simple line-fitting methods. School Psychology Review, 21(2), 300–312. doi:10.1080/02796015.1992.12085616


