Author ORCID Identifier

Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Education Sciences

First Advisor

Dr. Jennifer Wilhelm


This mixed methods study compared two groups of high school students’ understanding of the ionic bond and the dissolving process. A 5 lesson curriculum unit was developed using Taber’s electrostatic framework (1997) focusing on the electrostatic forces between ions compared to a molecular framework (business-as-usual) and the Next Generation Science Standards (NGSS Lead States, 2013). The lessons were developed to integrate spatially integrated experiences. Experimental (new curriculum unit) and business-as-usual (criss-cross method) students had their spatial skills tested before and after learning about the ionic bond using the Purdue Spatial Visualization-Rotations Test (PVST-Rot; Bodner & Guay, 1997). Students’ content was tested (pre and post) using the Chemical Bonding and Dissociation Diagnostic Assessment (CBDDA; Jang, 2003; Tan & Treagust, 1999; McBroom, 2011). Part of the assessment had two-tiered multiple-choice questions. Another part focused on dissolving of ionic compounds in water (dissociation equations and drawing ionic compounds dissolved in water). This study had one group of students using the new curriculum unit focused on ionic crystals, and the second group used more traditional methods of lab plus lecture.

Research Question 1: How does the understanding of the ionic bond and dissolving of ionic compounds in water compare for students using a unit focused on an electrostatic framework to students utilizing a molecular framework (business-as-usual) related to their spatial ability and using the spatial ability as a covariant with treatment group?

Research Question 2: How do spatial visualization skills compare for students using an electrostatic framework and students focused on a molecular framework?

A model using multiple linear regression was developed for Research Question 1 with the post score for the CBDDA as the dependent variable with pre score on the (CBDDA), Treatment, PVST-Rot Gain (post score minus pre score), and Treatment * PVST-Rot Gain as the independent variables. The null hypothesis was rejected, F(4, 87) = 4.674, p < .05. The model shows statistically evidence that it may predict the score on the post content test. Only the constant and the treatment group were the only statistically significant slopes.

Multiple linear regression was used to develop a model using the pre PVST score and the treatment group as the independent variables with the post PVST score as the dependent variable. The null hypothesis was rejected for Research Question 2, F(2,89) = 26.732, p < .05. Only the constant and pre PVST score slope were significant.

The qualitative portion of this study utilized the following sources: pre and post student interviews, drawings and dissociation equations from the CBDDA, classroom observations, and teacher logs. Some experimental students were able to improve their dissociation equations and/or the drawings compared to the business-as-usual group. The business-as-usual teacher logs reflected a more molecular framework of teaching. Interviewed students from both groups showed a lack of understanding of the difference between covalent and ionic bonding. Students from both groups did not comprehend that a molecule is used for only covalent compounds.

The approximate 3.2 experimental students to one business-as-usual student may be a limitation of this study. The new unit has potential to aid with the understanding of the ionic bond and the dissolving process.

Digital Object Identifier (DOI)

Funding Information

Arvle & Ellen Turner Thacker Research Fund, University of Kentucky, 2018