Group+Two+March+12

Article One: (Clarice, Sarah, Pam, Nicole)

**Spires & Donley (1998) Discussion**

 * I. RATIONALE FOR THE STUDY**
 * This study (1998) investigated the implementation of a prior knowledge strategy (PKA), as well as a combined main idea (MI) and PKA strategy. Through two separate experiments, the researchers sought to test the PKA and MI-PKA strategies and the effect of their use on students' performance on literal and higher-level thinking questions. The main reason for the study was to test a hypothesis that "instruction in this prior knowledge activation strategy would enhance ninth-grade students' comprehension of informational as well as literary texts" (p. 250). The researchers positioned their study as an extension of the known knowledge about prior-knowledge reading strategies. The researchers believed that viewing reading through "the lens of personal experience and exploration" (p. 250) could encourage student engagement and create a desire to stick with reading. They wanted to compare the effects of a learned reading strategy (one that drew on personal and background knowledge) with a more efferent (Rosenblatt, 1989), or factual, approach to reading.**

This seems to be a quasi-experimental, longitudinal study because though students had random assignment, there is no random sampling of a population; participants were enrolled in social studies classes at a particular urban high school, and results from year one were compared with results from year two with different students. They were 112 ninth graders, divided equally as low-, average-, and high-level readers as measured by the California Achievement Test and randomly assigned to three treatment groups through stratified random sampling, meaning there was a mix of reading levels in each group. Demographics aren't explicitly stated, but it can be inferred that the demographics of these groups are the same as is in the school. Independent variables were the three different treatments: 1) Prior Knowledge Activation (PKA), 2) Main Idea Strategy (MI), and 3) a no-instruction control group. The dependent measures included four tests - three immediate and one delayed - which were scored by two raters. Students who missed more than one day of training or who were absent during any of the testing were excluded from the study, leaving a total of 79 participants.
 * II. CRITIQUE OF RESEARCH METHODS**

In the second study, 170 ninth grade students were studied, using the same procedure - three groups and final tests.

One issue of internal validity was having researchers conduct the instructional treatments, but this was addressed. Having two raters addressed the issues of internal validity/reliability, and having multiple-choice tests were used (reliability). Content validity was addressed by having social studies teachers and a reading specialist read all passages and corresponding tests. The researchers acknowledged that they were unable to control for teacher effects. This study could be generalized to other populations in a similar manner.


 * III. CRITIQUE OF DATA ANALYSIS**

Each group of students was administered a total of four reading tests. The groups were asked to utilize the skills taught to them previously when answering the test questions. Because the tests combined both multiple choice and open-ended responses, the researchers chose to use two raters to score the open-ended responses. These scorers followed strict criteria and scored each open-ended response on a 0-3 point system. The tests and their questions were clearly designed in correlation with the research questions and thus, are appropriate for the study.

The researchers applied an inferential statistical approach to analyze the data. The researchers clearly point out that because there was only a "marginally significant difference across groups", they felt the need to utilize a covariate. *Dr. Spires, can you explain this process in depth tomorrow? I know p value should be .05, but I would like to hear about this process in action to clarify my understanding! :) After running further tests, the researchers were able to determine a significant difference for the treatment groups.

When student data was analyzed for literal comprehension, the Main Idea Strategy group out-performed both the control and Prior Knowledge Activation groups. When the data was analyzed for application-level comprehension, the students who comprised the Prior Activation group out-performed both the control group and the Main Idea Strategy group. Both groups receiving the treatment out-performed the control group in the area of multiple-choice responses. After clearly explaining their process, the researchers are correct in determining that explicit instruction on the application of prior knowledge when reading clearly impacted student scores.

The data was certainly triangulated when researchers chose to do a second study that combined instruction of both the PKA and MI strategies for reading. The data analysis for these subsequent test results was clearly described. Upon completion of the data analysis, the researchers determined that the groups activating PKA (both PKA and PKA-MI) had increased positive attitudes about reading.

IV. CRITIQUE OF DISCUSSION


 * In the discussion of experiment 1, the researchers concluded that PKA group performed consistently well on the open ended and multiple choice questions and speculated that students may have felt more invested in the reading task. Even though students in the PKA group performed better on the whole, they did not have an advantage in basic literal comprehension. The MI group had a slight advantage in literal comprehension. This caused the researchers to reevaluate and extend their hypothesis. The design of the second experiment reflected and inductive line of thought with a new hypothesis and change in instructional design so teachers conducted the research. This shift in design increased internal validity in the second experiment. **


 * The study connects its findings to the widely-accepted theory that activation of prior knowledge increases student engagement and comprehension; however, it extends that theory by encouraging students to continuously and recursively connect the text to prior knowledge rather than in a pre-reading lesson. It is also different from many previous studies regarding prior knowledge because it focuses on informational text rather than narrative. **


 * Spires and Donley suggest that future research address "students' thinking processes as they engage in an aesthetic stance with informational text" and "explore ways to incorporate instruction...in a more natural context." (p. 258) **

V. ADDITIONAL OBSERVATIONS (QUESTIONS/CONFUSIONS??) Is the difference between random assignment and random sampling as mentioned above the fact that the students selected from the school were not randomly pulled from various classrooms? I am confused as to the difference.

I am also wondering, even though the students were instructed to use the PKA, MI, or 'silent reading' strategy they were taught, how would the researchers know that the students ACTUALLY used those strategies for the test? We always tell kids how to do various strategies for the EOC or other tests, but there's no way to know if they actually use those strategies in a testing environment.**