PROJECTS

Explore below to learn more about our work on Early Reading Assessments. More details coming soon regarding our work on Academic Interventions and Collaborations and Partnerships. Stay tuned!
Assessments of Reading Acquisition
Educators use academic data to make important decisions about students educational programming every day. Whether they are low stakes decisions, involving students’ classroom experiences, or high stakes decisions that involve long-term decisions about students’ abilities, placements and progress, tests used to make these decisions must provide evidence that support valid conclusions. PAART investigates whether commonly used reading assessments meet psychometric standards of reliable testing procedures, elicit valid results and are fairly interpreted to improve students’ experiences in schools.
Current Projects
These are the questions we are currently investigating:
1. Can the skills of reading acquisition be effectively measured using the screening measures that are commonly available in a way that models developmental reading theory?
-
Does early phonological awareness predict early reading acquisition?
-
Do early decoding skills predict reading fluency?
-
Is fluent passage reading really predictive of later reading comprehension?
-
Are these early literacy skills prerequisites to later reading development?
Using a longitudinal sample of approximately 270 students from two district-wide cohorts, we will test whether early reading skills are predicted of the acquisition of later reading skills from Kindergarten through third grades.
Completed Projects
Examining the Predictive Validity of the Developmental Reading Assessment
The purpose of this study was to examine the predictive relationships between students’ performance on the Developmental Reading Assessment (DRA2) in the early years of their reading acquisition and their later reading achievement in third grade as measure by the Stanford Achievement Test- Tenth Edition (SAT-10). Data were obtained from 179 students who attended a public charter school for grades kindergarten through third grade during the years 2008 to 2015.
We Predicted:
-
The DRA2 would not be an effective measure for predicting later reading achievement when used in kindergarten and first grades, because it was not designed to measure phonemic decoding processes of important for emergent readers. However, we expected that it would be an effective measure for use in second grade.
-
Word study and word reading skills measured in kindergarten and first grade would account for a significant proportion of the variance in reading comprehension in first and second grades, respectively, and that the DRA2 would not significantly improve the predictions.
-
Once phonemic decoding was effectively attained in second grade, the DRA2 would be a significant predictor in a model that included word study and reading vocabulary.
We Learned:
-
The DRA2 moderately predicted later reading comprehension across all grade levels, and the predictive relationship grew stronger across the years (r = .60 from kindergarten to first grade, r = . 63 from first to second grade, and r = .68 from second grade to third grade).
-
The kindergarten DRA2 did not account for a significant and unique proportion of the variance in first grade reading comprehension when phonemic decoding skills were accounted for.
-
The DRA2 was a significant predictor in the models for assessing reading in first grade, and consistent with our hypotheses, was also significant for assessing second grade reading to predict third grade comprehension.
This study was presented at the Northeastern Educational Research Association Meeting in Trumbull, CT.
Evaluating Validity Evidence for One Informal Reading Inventory
Using the sample from our previous predictive validity study, the purpose of this study was to evaluate whether one IRI (the DRA2) elicits scores from which valid decisions can be made. Using Receiver Operator Curves, we tested whether scores from the DRA2 can be used to reliably determine which students may be at-risk for later reading failure. The data used in these analyses included 179 students who attended a public charter school for grades kindergarten through third grade during the years 2008 to 2015.
We Predicted:
​
-
The DRA2 would not accurately identify students who are at risk for reading problems when they are used in Kindergarten or First grade, but would have better classification accuracy when used with second grade students.
We Learned:
-
The AUCs were larger across all grade levels than we had predicted.
-
As we predicted, the DRA was not useful for identifying the Kindergarten children who were at risk for later reading comprehension failure.
-
More accurate decisions were observed using the first grade DRA2 screener than were observed for the Kindergarten and the second grade screener.
-
The common decision error across each grade is the sensitivity of the DRA2 to accurately identifying at-risk readers. These results are consistent with findings from similar IRIs in previous studies (Klingbeil, et. al, 2015.)
This study was presented at the annual conference of the National Association of School Psychologists in New Orleans, LA. (See Poster)
Academic Interventions
More coming soon!
Collaborations & Partnerships
More coming soon!