Options
Singapore post-secondary students’ sense-making of statistical information in graphs
Loading...
Type
Thesis
Author
Chia, Hui Teng
Supervisor
Ng, Swee Fong
Abstract
Graphs are widely used to communicate statistical information. Research shows that school children, students and even professionals are challenged to provide accurate and meaningful interpretations of the information in graphs. Various factors, such as statistical beliefs, habits of mind and knowledge, have been offered to try to explain such challenges. The purpose of this study is to examine the interplay of these factors to explain how people make sense of statistical phenomenon in graphs.
The corpus of empirical evidence came from two phases, Phase 1: Written Tasks and Phase 2: Task-based interviews. Phase 1 provided the initial insights into participants’ statistical sense-making components of beliefs, habits of mind and knowledge. Participants were required to respond to two paper-based tasks: the Bar Graph Task and the Line Graph Task. The Structured of the Observed Learning Outcome (SOLO) taxonomy and qualitative classification of responses were used to analyse participants’ written responses.
In Phase 2, participants were interviewed as they performed two tasks, the Sorting Task and the Labelling Task. Both the interview tasks were designed based on the theory of discernment. In the Sorting Task, participants were asked to sort 12 graphs into categories they deemed appropriate. They were then asked to discuss the reasons for these categories. How the participants sorted the graphs revealed whether participants discerned surface features or deep features of the graphs. Surface features included explicit features such as the colours, types of graphs, keywords in the title and so forth. Deep features required some abstractions of the specifiers in the graphs such as the stories that participants could tell with the information presented in the graphs. The participants’ discussions of the graphs were interrogated for their statistical sense-making components of beliefs, habits of mind and knowledge.
The Labelling Task teased out participants’ statistical knowledge on the important notions in statistics such as ‘sample’, ‘population’, ‘randomness’, ‘representativeness’ and ‘variation’ as well as their knowledge of the sources of statistical data. Participants were required to select some labels to discuss two line graphs. The responses to Labelling Task were analyzed and arranged into a hierarchical structure to show the levels of sophistication in statistical knowledge.
Eighty-eight post-secondary diploma students from one polytechnic in Singapore participated in Phase 1 of the study, of which 22 also participated in Phase 2.
The findings from Phase 1 and Phase 2 of this study were consolidated to construct the Diversity in Sense-Making framework. This framework is captured on a spectrum, with Expert-like sense-making at one end and Novice sense- making on the other. The sense-making of statistical phenomenon in graphs is shown to be a function of statistical beliefs, habits of mind and knowledge. The interplay between these three components is captured by segments in a doughnut-like ring. The different ways the segments in a doughnut-like ring interact were associated with the different ways the participants of this study interpreted statistical information, meaningful or otherwise, in graphs. The framework is used to suggest how to advance the field further.
The corpus of empirical evidence came from two phases, Phase 1: Written Tasks and Phase 2: Task-based interviews. Phase 1 provided the initial insights into participants’ statistical sense-making components of beliefs, habits of mind and knowledge. Participants were required to respond to two paper-based tasks: the Bar Graph Task and the Line Graph Task. The Structured of the Observed Learning Outcome (SOLO) taxonomy and qualitative classification of responses were used to analyse participants’ written responses.
In Phase 2, participants were interviewed as they performed two tasks, the Sorting Task and the Labelling Task. Both the interview tasks were designed based on the theory of discernment. In the Sorting Task, participants were asked to sort 12 graphs into categories they deemed appropriate. They were then asked to discuss the reasons for these categories. How the participants sorted the graphs revealed whether participants discerned surface features or deep features of the graphs. Surface features included explicit features such as the colours, types of graphs, keywords in the title and so forth. Deep features required some abstractions of the specifiers in the graphs such as the stories that participants could tell with the information presented in the graphs. The participants’ discussions of the graphs were interrogated for their statistical sense-making components of beliefs, habits of mind and knowledge.
The Labelling Task teased out participants’ statistical knowledge on the important notions in statistics such as ‘sample’, ‘population’, ‘randomness’, ‘representativeness’ and ‘variation’ as well as their knowledge of the sources of statistical data. Participants were required to select some labels to discuss two line graphs. The responses to Labelling Task were analyzed and arranged into a hierarchical structure to show the levels of sophistication in statistical knowledge.
Eighty-eight post-secondary diploma students from one polytechnic in Singapore participated in Phase 1 of the study, of which 22 also participated in Phase 2.
The findings from Phase 1 and Phase 2 of this study were consolidated to construct the Diversity in Sense-Making framework. This framework is captured on a spectrum, with Expert-like sense-making at one end and Novice sense- making on the other. The sense-making of statistical phenomenon in graphs is shown to be a function of statistical beliefs, habits of mind and knowledge. The interplay between these three components is captured by segments in a doughnut-like ring. The different ways the segments in a doughnut-like ring interact were associated with the different ways the participants of this study interpreted statistical information, meaningful or otherwise, in graphs. The framework is used to suggest how to advance the field further.
Date Issued
2018
Call Number
QA276.3 Chi
Date Submitted
2018