Student-centered Assessment helps students examine their own learning. Effective feedback on performance--and assistance on how to reflect on it--encourages persistence. So give students feedback they can productively use, i.e., timely, contextualized, and actionable.

Some suggestions

Provide frequent opportunities for feedback. Students need to understand how they are doing in relation to their peers, to your expectations, and to their final grade. Instead of having students wait until a midterm to gauge their performance, try giving short knowledge quizzes on previously-covered material. Clickers are a great way to do this.

Help students interpret the results of any assessment. This begins with simply reporting the mean and standard deviation for every assessment. Then let students know what is considered "doing well" in the class, and what they can do if their grade is below that (e.g., come to office hours, attend an extra lab, meet with your TA).

High performing students need to understand their performance as much as struggling students. Sometimes we are so focused on helping students in trouble, that we don’t give enough feedback to those who are succeeding. Don’t hesitate to drop an email to a student who is doing well letting them know that you noticed. This kind of recognition from a professor can be transformative, especially to a student who may feel like they don’t belong.

Encourage students to seek help. When a student struggles, encourage them to persist through the task, and make it acceptable and easy to seek help. For example, periodically remind students when and where they can get help and genuinely encourage them to take advantage of it.

Examples from the collection

Twitter Trends

In this project, designed to be completed in teams of two, students solve twelve problems that incrementally will have them develop a program to visualize positive/negative sentiments of Twitter posts (tweets). The twelve problems are organized into four distinct phases. The problems require that students either apply and/or implement an abstract data type (ADT). For example, students implement ADTs for tweets and sentiments. The problems require using ADTs like lists and dictionaries.

Engagement Excellence

Cluster Analysis

Gives students practice with writing classes using k-means clustering, a common machine learning technique. Students will be exposed to classes with complex invariants and practice using data encapsulation in classes. This is a challenging assignment but it is sure to be a good learning experience for students.

Engagement Excellence

Seam Carving

This is an assignment on implementing an algorithm that generates a context-aware image resizing technique. Students work to create a program that identifies the path of pixels connected from either a vertical or horizontal seam in one pixel increments to generate a resized image that preserves interesting features (such as aspect ratios and set of objects present). Mutable data types must be implemented to identify seams and resize the image.

Engagement Excellence