1 of 47

Providing Educational Insights Using AI

DR. LAURA SAMULSKI-PETERS

2 of 47

Why AI?

  • - Manual data review is time-consuming
  • -Teachers and Administrators are not trained data analysts
  • - Bias in interpretation (Although AI can also have biases)
  • - Difficulty identifying complex patterns
  • - Limited predictive capabilities

3 of 47

How AI Can Help

- Automation: Speeds up data processing

- Pattern Recognition: Finds trends not easily visible

- Predictive Analytics: Forecasts future incidents

- Bias Detection: Highlights disparities by race, gender, etc.

4 of 47

Ethical Considerations

  • - Data privacy and student confidentiality
  • - Avoiding algorithmic bias
  • - Transparency in AI models
  • - Human oversight is essential

5 of 47

Getting Started

  • - Collect clean, structured data
  • - Partner with data scientists or use AI tools
  • - Start with descriptive analytics, then move to predictive
  • - Train staff on interpreting AI insights

6 of 47

Three Interesting Use Cases

  • Streamlining the DDI Process
  • Descriptive Analytics
  • Predictive Analytics

7 of 47

DDI

Assess

Analyze Data

Develop Action Plan

Implement/Monitor Action Plan

Revise the Plan as Needed

8 of 47

Let’s see how AI does with DDI!

Examine the iReady Math data on the next slide.

As a table make three observations about the data.

Make sure someone writes your three observations down.

9 of 47

Remember…

OBSERVATIONS ARE FACTUAL STATEMENTS ABOUT THE DATA THAT INCLUDE NO EXPLANATION!

10 of 47

Race/Ethnicity

3 or More Grade Levels Below

2 Grade Levels Below

1 Grade Level Below

Early On Grade Level

Mid or Above Grade Level

Asian

91

69

122

9

1

Black/African American

327

276

196

10

0

Hispanic

186

119

93

4

2

Multiracial

20

33

34

0

1

White

53

80

138

25

10

11 of 47

12 of 47

13 of 47

14 of 47

15 of 47

How many of your observations were on the list?

16 of 47

17 of 47

18 of 47

19 of 47

20 of 47

21 of 47

22 of 47

23 of 47

24 of 47

Descriptive Analytics

I created a large file of grade 3 Mathematics performance over a year. The data included race/ethnicity, gender, sped status, ell status, formative assessment data (BOY, MOY, EOY), student ADA, NYS grade 3 assessment level, final grade and teacher.

25 of 47

I also asked if there was a statistically significant difference between grades by teacher…

26 of 47

27 of 47

28 of 47

29 of 47

30 of 47

31 of 47

32 of 47

33 of 47

34 of 47

35 of 47

Predictive Analytics

36 of 47

37 of 47

38 of 47

39 of 47

40 of 47

41 of 47

42 of 47

What else should we ask?

43 of 47

One last example for my Qualitative friends!

44 of 47

45 of 47

46 of 47

47 of 47

Thank you!

LSAMULSKI-PETERS@BUFFALOSCHOOLS.ORG