Dana Center design principles: active learning

eCOTS 2022 Workshop

Nicholas J. Horton and Benjamin S. Baumer

May 19, 2022

Dana Center Data Science Design Principle Framework

• The framework is copyright 2021 The Charles A. Dana Center at The University of Texas at Austin.
• The Dana Center has granted educators a nonexclusive license to reproduce and share copies of this publication to advance this work.

Design Principle: Active Learning

The course provides regular opportunities for students to actively engage in data explorations using a variety of different instructional strategies (e.g., hands-on and technology-based activities, projects, small group collaborative work, facilitated student discourse, interactive lectures).

Design Principle: Active Learning

The course provides regular opportunities for students to actively engage in data explorations using a variety of different instructional strategies (e.g., hands-on and technology-based activities, projects, small group collaborative work, facilitated student discourse, interactive lectures).

Student perspectives

• Be active and engaged participants in discussion, in working on data explorations with classmates, and in making decisions about the direction of instruction based on their work.
• Discuss results of their data explorations with the instructor and/or classmates in class.
• Develop and evaluate data- based arguments.

Student perspectives (cont.)

• Think critically about data and be open to changing their mind after considering data-based arguments presented in class.
• Consider the implications of their conclusions within the context and as part of a broader picture, including consideration of data ethics.

Faculty perspectives

• Provide low-floor, high-ceiling activities and explorations that all students can access and that extend to high levels. Such activities should provide meaningful opportunities for exploration and co-creation of mathematical understanding and data literacy.
• Engage students through relevant contexts by providing local data sets and inviting students to ask questions about the data. Encourage different students to pose and investigate different questions that can be addressed by exploring data, and to come together to discuss findings.

Faculty perspectives (cont.)

• Ensure that all students are provided equitable opportunities to engage successfully throughout the course.
• Facilitate students’ active learning of data science through a variety of instructional strategies, including inquiry, problem solving, critical thinking, and reflection.

Faculty perspectives (cont.)

• Create a safe, student-driven classroom environment in which all students feel a sense of belonging to the class and the discipline, are encouraged to take risks and embrace mistakes, and are able to make decisions about the direction for instruction through the results of their exploration of data science. Students’ ideas are at the center of the conversation.

Small group discussion

Please take four minutes in groups to briefly discuss how you have incorporated Active Learning into your courses.

The group leader (person whose first name appears first in the alphabet) is asked to share one or two examples that arose in the discussion which resonated for the group.

How to operationalize?

Case study 1: Leaflet maps

• introduce via simple example (start of class period)
• point students to the Leaflet for R page: https://rstudio.github.io/leaflet
• have them add an element to a simple leaflet map (during class)
• have them work in groups to create a dynamic map using a new dataset (outside of class)
• see leaflet.Rmd

Case study 2: Teaching regular expressions (finding patterns in text)

• introduce via short lecture (previous class period)
• have students read a short section or watch video to provide a framework for regular expressions as a langauge
• use time in class to let students work in groups on a worksheet with multiple examples of regular expressions
• homework assignment to provide additional practice

Examples (in class)

x <- c("popular", "popularity", "popularize", "popularise", 
      "Popular", "Population", "repopulate", "reproduce", 
      "happy family", "happier\tfamily", " happy family", "P6dn")
grep(pattern = 'pop', x)  #1 

[1] 1 2 3 4 7

grep(pattern = '^pop', x) #2

[1] 1 2 3 4

grep(pattern = 'populari[sz]e', x) #3 

[1] 3 4

grep(pattern = 'pop.*e', x) #4

[1] 3 4 7

grep(pattern = 'p[a-z]*e', x) #5

[1]  3  4  7  8 10

grep(pattern = '^[Pp][a-z]+.*n', x) #6

[1] 6

Examples (in class or homework)

How often does the word “science” appear in Emily Dickinson’s poetry?

SOLUTION:

#remotes::install_github("Amherst-Statistics/DickinsonPoems")
library(DickinsonPoems)
library(tidyverse)
library(tidytext)
all_poems <- list_poems() %>%
  map(get_poem) %>%
  unlist() %>%
  enframe(value = "words") %>%
  unnest_tokens(word, words)
head(all_poems)

# A tibble: 6 × 2
   name word     
  <int> <chr>    
1     1 i        
2     3 success  
3     5 published
4     5 in       
5     5 a        
6     5 masque   

Solution (cont.)

all_poems %>%
  filter(stringr::str_detect(word, "science"))

# A tibble: 4 × 2
   name word      
  <int> <chr>     
1   511 conscience
2  4618 science   
3  6549 sciences  
4  7849 science   

See also text.Rmd

Some caveats and notes

• It’s often helpful to provide the students with a set of notes about regular expressions as they work through their exploration (see RStudio cheat sheet on “String manipulation”)
• It’s very useful to start very simply then build in new bells and whistles slowly (see Growth Mindset)
• More sophisticated aspects of regular expressions can be divided into groups, where each group presents their solution to the class as a whole (and shares the results in a class Slack channel for future reference)