"They Know, but They Don't Know": Expanding Relationality Through Hip-Hop Making

• In Event: Spit What You Know: Critical Notions of Knowing, Showing, and Educational Assessment Through Hip-Hop

Abstract

Objectives
We present a design-research performance task that examines relationships between youth cultural and computational practices. Our larger projects seeks to develop learning ecologies grounded in the five elements of hip hop (Akom, 2009), culturally sustaining pedagogy (Paris & Alim, 2017), and computational making (Authors, 2021a) to study relationships between STEM learning, hip hop, and identity (Authors, 2020). Youth addressed the question “can you teach a computer to DJ?” by interrogating algorithmic approaches to playlist construction. We share this performance assessment of computational activity and related findings.

Theoretical Framework
Culturally sustaining approaches to assessment disrupt persistent “adultism” (Bertrand et al., 2020) that dominates modes of measuring knowledge. Assessment culture operates with an “ideology of adult supremacy” (DeJong & Love, 2015, p. 490), where youth knowledge is “measured” against objectified constructs. Critical Hip Hop Pedagogy (CHHP; Akom, 2009) situates youth relational practices, like DJing, as forms of knowledge production (Nasir & Hand, 2009) to bridge conversations between CHHP and efforts to learn what students know.

Methodology
We developed a playlist assessment task in collaboration with a community education organization centering hip hop production, performance, and history (Authors, 2021b). Youth built 10-track playlists from a common seed track. Groups discussed their approaches, then assigned “tags” to each song based on their approaches--creating a parallel with how computational platforms generate playlists automatically based on objectified categories. We ask “What practices do youth exhibit in culturally sustaining assessments of computational systems?”

Data
Data include transcribed Zoom recordings, fieldnotes, and interviews with high schoolers. We use case-study methodology (Stake, 1995), to address the research question, sharing how two youth constructed a playlist and assigned “tags.”

Findings
We present a case of Chandler and Royalty using relational constructs to build a playlist. They selected tracks “by mood or what sounds good together,” but also assigned descriptors like “West Coast” and “Medium Temp” to tracks like California Love by Tupac and It Was a Good Day by Ice Cube. The youth used categorizations much like Pandora’s use of “genes”, however, they also challenged the use of objectified attributes (Tingle et al., 2010). Chandler said:
“I learned, it’s kind of hard for me to put something that’s more logical … I think about coast, region, tempo as like more … you can collect that as data. But when I think about that, or what song reminds of what song, it’s usually about feeling. So, thinking about what you can gauge as far as real “yes or no” answers … is a cool way to think about it, which I hadn’t really thought about before.”
The case illustrates how youth relational categorizations supported their learning about computational algorithms.

Significance
The relational nature of youth practices in playlist construction stands in contrast with objectified approaches used by algorithm-based platforms like Pandora. Viewing assessment through a CHHP lens challenges the premise of measuring knowledge against objective, “adultist” constructs. Culturally sustaining tasks hold space for youth to express their ways of knowing and doing in relation to the learning goals of the project.

Authors

• Brian E. Gravel, Tufts University

• Eli Tucker-Raymond, Boston University

• Amon D. Millner, Franklin W. Olin College of Engineering

• Samara Fair, Boston University

• Adriana Alvarado, Boston University