Edgar, J. W. (2005). The Effect of Computer Based Remediation on Chemistry Students Achievement as Measured by Benchmark Assessments . Instructional Technology Monographs 1 (1). Retrieved <insert date>, from http://itm.coe.uga.edu/archives/spring2005/filename.htm.

The Effect of Computer Based Remediation on Chemistry Students Achievement as Measured by Benchmark Assessments

by

John W. Edgar
University of Georgia

 

Abstract

College prep chemistry classes were exposed to computer based remediation activities in an effort to determine the effectiveness of computer based tutorials. Benchmark assessment of key objectives were given before and after the remediation activities. The study showed an increase in achievement on the post test for the majority of students and student surveys indicated a positive student response to using computer based activities designed by the teacher.

 

Literature Review Methods Results and Discussion Conclusions References

 

Introduction

The last 15 years have witnessed the emergence of two major players in secondary education. The increases in functionality and affordability of computers along with the development of the Internet have triggered many schools to purchase computers in an effort to integrate technology. Computers are more prevalent in schools today than ever before and the use of the Internet by high school students has increased dramatically. Within this same time frame, many states have developed and implemented minimum competency tests/graduation tests that are considered high risk/high stakes tests because student have to pass them in order to graduate or move on to the next grade level. The NCLB Act of 2001 mandates that states monitor growth of learning in addition to ensuring students have attained minimum competencies in core subjects. The advent of these test and the NCLB Act of 2001 puts a greater emphasis on testing in schools and therefore demand! s that teachers make every attempt to make their students successful. The time constraints present in education make it important that teachers find ways to check their students’ progress and develop efficient methods of remediation.

Gwinnett county, a metropolitan suburb of Atlanta, has started a program that will design and implement benchmark assessments at 6, 12, 24, and 30 weeks during the school year. These tests are being designed to ensure that students have achieved mastery of basic objectives (termed academic knowledge and skills AKS). Along with pre and posttests, the benchmark assessments intend to show student growth within classes. The benchmark assessments are constructed in the same vein as high stakes graduation tests. Both are multiple-choice tests that are aligned with standards established by the state. Debate about the validity of the tests is prevalent in research literature, yet discussions about remediation for students who do not pass are almost nonexistent. The emerging technology associated with computers and the Internet is an untapped tool that could be utilized to increase test scores and provide an efficient remediation process. Studying the ef! fects of this type of remediation when used with benchmark assessments will allow educators to determine if the remediation process will be effective with graduation tests. However, while the benchmark assessments will provide teachers direction for remediation with their individual students feedback from current graduation tests is not specific to individual students. At least two states are implementing online testing for graduation tests that will eventually provide specific information to teachers and thus allow the teacher to direct remediation (Neugent, 2004; Woodfield, 2003).

Literature Review

Research Methods

The review of the literature was conducted using Galileo, primarily searching through Academic Search Premier, ERIC, and Education Abstracts databases. Articles dealing with remedial courses at higher education levels were found most when using remediation as a search term. Upon narrowing the search to secondary education few results were found. Tutorials as a search term yielded more results but few related to graduation tests. An abundance of articles were found concerning the validity and effect of graduation tests. Additionally many articles were found concerning the use of computers and technology in the classroom but again few relating to the remediation of students with failing scores on graduation tests. References from several articles were used to find other studies but several were not available online. The Google search engine was used to find titles of programs and links to tutorials of which there were many. Finally the author use! d personal experience and discussions with colleagues especially with reference to the Georgia High Graduation and the current development of benchmark assessments in Gwinnett county Georgia.

Review Questions

This literature review was conducted to provide the author with the necessary information to make informed decisions with regard to the remediation process to be used by teachers. Specifically the review attempted to answer the following questions:

  1. What types of remediation have been used with students?
  2. Have technology based remediation tools been effective?
  3. Has the use of similar technology be effective in the classroom?

The types of remediation are broken down into two different categories: Traditional and Nontraditional. Traditional types of remediation involve teacher-student contact or student-student contact. These typically occur within the setting of the school the student attends. Nontraditional types of remediation involve student-computer contact or student-internet-teacher contact. These types of remediation can occur anywhere students can gain access to the programs. If nontraditional types of remediation are effective, they should also be used to supplement or change instruction in the classroom. The author has attempted to alter instruction with different types of technology integration but realizes there are more possibilities to be explored. Thus the review also looked for classroom application involving nontraditional applications. The impetus for the review came from a desire to improve student achievement on graduation tests. While not a dire! ct question, investigation into these tests and other related tests would help shape future research questions.

Traditional Remediation

Traditional types of remediation in the secondary setting usually involve students spending more time with staff after regular school hours. The student has the choice of staying after school, coming to school on Saturdays or attending summer school. While some students experience success with these opportunities they are being exposed to “the same learning activities using the same texts and workbooks with the same ineffective teaching” (Starratt, 2003) . Within these experiences students could be exposed to remediation of basic concepts, working of example problems, bridging techniques and student discussion groups (Mason & Crawley, 1993).

Nontraditional Remediation

Nontraditional types of remediation can include software programs or online tutorials. While a google search will show several examples of both types, journal literature does not provide much research on their effectiveness. Anderson (Anderson et al., 1995) details early work in designing “cognitive tutors” that attempt to tutor students with computer programs. The Advanced Computer Tutoring Project at Carnegie Mellon University appears to be the first step in using computers as tutors. While designers of these and subsequent programs and software packages attempt to use constructivist practices, most applications surveyed fell into the category of “drill and practice” (Morse, 1991; Nakhleh et al., 2000; Schneider, 2000a, 2000b; Scott, 2001a, 2001b; Sherman, 1999; Tillman, 2000; Timmons, 2001) . While “drill and practice” may have a niche in education they do not have the flexibility to allow! students to construct knowledge.

Online tutorials do not differ much from non-online tutorials, they are however more readily available and less expensive. The Georgia State Department of Education offers tips and practice tests online that will score student attempts. Using this service gives students an idea about the type of questions that will be used, but the knowledge gained is strictly factual and does not tie together major concepts. Several applications have been developed to complement post secondary classes. While students find them more flexible, there is no research to show that they improve student test scores (Carswell et al., 2000; Crowther et al., 2004; Littlejohn et al., 2002).

Nontraditional types of remediation offer students more flexibility and different instructional methods not found in their courses. A large number of students can benefit from this flexibility as they participate in programs outside of regular school hours or summer school. However a large percentage of these students do not have appropriate access to computers and the Internet away from school. Unfortunately this segment of the population typically scores lower on standardized tests. Gains in achievement can be demonstrated with both types of remediation but tutorial programs have not succeeded in teaching high-level reasoning (Horwitz, 1999) . In order for students to truly benefit from a remediation program they must have appropriate access to interventions that increase their knowledge base as well as their capability to learn.

Technology Use in the Classrooms

Computer applications for the classroom come in a variety of shapes and sizes each with their own twist on teaching. Typically teachers have looked for openings into which they could insert an application rather than change their curriculum. Older applications worked nicely with this model, as they were mostly objective drilling exercises that augmented lecture. Current philosophy in education advocates switching to a more constructivist approach, likewise most computer applications are attempting to become constructivist in design (Rodrigues, 2000) . While most computer applications have the capability to increase student achievement, (Christmann & Badgett, 1999) there is support that constructivist applications would actually enhance student learning. Constructivist software would increase student metacognition and develop life long learners (Horwitz, 1999; MacKinnon, 2001; Windschitl, 2000) . Usage of certain applications and the Internet! can easily be incorporated into project-based learning and problem-based learning activities (Churach & Fisher, 2001; Cox-Petersen & Olson, 2000) . While exciting and innovative this approach to teaching (especially with the use of technology) does not match the format of a typical graduation test. Teachers face a dilemma when planning a lesson: Do I use tools to make my students better learners or do I focus my attention on preparing my students to pass a high stakes test.

Graduation Tests

The USA Today reported the 25 states have high school graduation tests (2004) . The advent of the NCLB act of 2001 will likely increase that number by 2010. These tests are affecting teachers and their teaching practices. In Massachusetts teachers are changing their instructional practices to help students graduate and improve their schools assessment scores (Vogler, 2002). Teachers in a successful (high test scores) school district in Ohio report that the imposition of state tests has increased workload created excess stress and decreased faculty morale (Kubow & DeBard, 2000). These teachers feel rushed to cover material and that they are teaching to the test. While the tests may focus the curriculum and provide meaningful incentives, the validity of the tests are questioned by parents and teachers and impede instruction that leads to higher-order thinking skills (Jacob, 2001; Reising, 2000) . If the tests are given to students during their! sophomore or junior years, what does this tell the student about the learning that could be achieved after the test? Proponents argue that the test will show teachers student weaknesses and allow time for remediation. Currently remediation efforts are focused more on passing the test rather than improving learning. Most teachers could identify the majority of students who will fail a graduation test using previous test scores and course grades. The argument is that the students could be identified earlier so that interventions could take place before the test rather than remediation after the test (Gibson, 1997). Early intervention programs in the primary grades may reduce large numbers of low achievers in upper grades (Neal & Kelly, 2002). The reality of the situation is that the tests are not going away and teachers and schools are going to bear the responsibility of improving test scores. Schools and teachers will also bear the public scrutiny when test scores do not i! mprove resulting in a label of a “needs improvement school.”

Research Questions

Upon review of the literature the author has formulated the following research questions:

  1. Will classroom implementation of computer based programs and Internet based activities reduce the number of students who fail benchmark assessments?
  2. Will use of nontraditional types of remediation improve benchmark scores of students who did not pass the initial benchmark?
  3. Will the use of technology improve the learning capabilities of students as measured by final grades?

The development of the questions was influenced by the review of the literature and by the experiences of the author in the fall of 2004. Two benchmark assessments (6 and 12 week) were given during the fall semester to 10 th grade chemistry. Students failed the assessments (below 70%) in large numbers and traditional remediation helped only a few students. While using benchmarks seemed to indicate an increase of final exam scores, grades for the course were lower than in previous years. Based upon the literature the remediation process should improve test scores and course grades.

Methods

In order to increase student performance on benchmark assessments I created and used review tools and tutorials with Inspiration, PowerPoint and the Quia.com website. The review tools were designed to enhance student knowledge on core objectives. The data collected demonstrated the effectiveness of the technology-based tutorials on student scores who retook the test.

Design/Data Sources

Using a quasi-experimental design I collected data from pretests, benchmark assessments, and retakes of benchmark assessments. These instruments consisted of multiple choice question tests based on core objectives. The number of questions on the tests ranged from 20 to 45 questions. The benchmark tests are produced by the school district and correspond to the mandated degree objectives.

Participants

The participants in this study attend a suburban high school in the southeastern United States. The high school has a population of 2000 of which 50% are Caucasian, 30% are Asian, 10% African-American and 10% Hispanic. All participants were enrolled in a college preparatory chemistry class. More than 5% of the participants were taking the class a second time. Only those students who decided to take the retest of the benchmark are included in the study.

Procedure

At the beginning of the semester students were given a pretest that covers 4 core objectives. This test sets a baseline of prior knowledge for the students. The teacher demonstrated the use of PowerPoint and Inspiration during the course of instruction. These tools were used to make review guides or tutorials for the core objectives. At the end of the first six weeks, the first two core objectives had been covered and the first benchmark assessment was given. Students can opt to retake the benchmark assessment after they have used the review guides and tutorials created to review the core objectives. One day after the 6 week benchmark assessment all students were taken to a computer lab and directed to the website Quia.com. Students used teacher created games and activities to review the core objectives. Students were encouraged to try Quia.com at home and at the library during lunch. The teacher also asked students to create new questions ! to be used in the games and activities at Quia.com. The next day students used laptops in the classroom to take practice chapter tests online that were produced by the publisher of the textbook used in the class. The students worked in pairs as they took the tests and emailed the results to the teacher. They could retake the online practice tests until they were satisfied with their grade. One week later these students were given an opportunity to retake the first benchmark assessment. Data was collected that compared the scores of the two tests. After the retest of the first benchmark has been scored students were surveyed and interviewed (see appendix 1) to determine the effectiveness of the process. The researcher evaluated the suggestions made by the students and modifications were made to be used for remediation of the 12 week benchmark assessment. Data was collected that compared the scores of the two tests.

Analysis

Mean differences in scores were calculated using an Excel spreadsheet. The averages of the pretests and posttests were compared to determine student gains in achievement. Rates of improvement were also compared with rates of improvement from last year’s data. However direct comparison of the data between both years presented a problem. The previous year students were required to take the retest if they scored less than 70% while those who scored higher than 70% were not given the option to retake the test. In the current year students had the option to retake the test regardless of score. Thus data from students who scored less than 70% this year was pulled from the main data set to provide a direct comparison. (see appendix 3) Information from the interviews were categorized into major themes. The survey results were tabulated to indicate the number of students for each possible response. (see appendix 1)

Role of the Researcher

As the teacher of the classes involved in the study, I administered the tests and collected all of the data and was responsible for the remediation. I will be interviewing the students and distributing the surveys to the students.

 

Results and Discussion

48 students attempted the benchmark after completing the remediation activities. The average score on the first benchmark for the 48 students was 67.7%. The average score of the same 48 students on the retest was 75.2%. The net increase in average was 7.5. Not every student showed an increase for pre to post test however. Only 32 out the 48 increased their scores while 5 made the same score and 11 showed a decrease in their score. (see appendix 2)

Those students who initially scored less than 70% showed an increase of 8.7%. Although the average from this group was lower than the overall group (75.2% to 69.0%) there was a slightly higher increase from pretest to posttest. 17 out of 28 students increased their scores while 4 made the same score and 7 showed a decrease in their score. (see appendix 3) Students from the current year also showed a greater pretest and posttest average. Last year's students averaged 54% on the pretest and 64% on the posttest. While the percent increase is a higher numerical value for previous students, current students had an equivalent increase in correct responses. The 6 week benchmark consisted of 25 questions while the prior benchmark consisted of only 20 questions.

The results of the written survey indicated that 90% of students thought that using computers were very helpful or somewhat helpful. However 75% of the students spent less than 30 minutes outside of class using the computers and 85% of the students completed 3 or less of the remediation activities.

Class discussion of the benchmark remediation activities was varied and only a few ideas came up in all the classes. Students unanimously liked the instant feedback received from the activities and they preferred teacher created remediation tools rather than activities created by the publisher of the textbook.

Conclusions

Integrating technology into my classroom has been beneficial to myself and my students. Using computer based activities has increased the interest and achievement of my students and promises to make my teaching style more efficient. However my classroom is not a controlled environment designed to prove that the modifications I make increases learning. There have been many outside influences that have affected the design of this project and therefore should be mentioned. The original timeline had to be adjusted to incorporate changes in the instructional calendar and the redesign of the 6 week benchmark assessment. The method of delivery of the benchmark also changed in the midst of data collection as the district made the benchmark assessments online tests. Students were given the 6 week benchmark as a paper and pencil test with a scanable answer sheet but those who took the retest did so using laptops and computer la! bs online. There was no control group as I wanted all my students to have the same opportunities. The only comparisons made were with respect to last years students. Students in this year's study achieved higher averages and equivalent gains in achievement. In short while the numbers show the project to be somewhat successful, I do not feel that definite conclusions should be drawn.

As mentioned previously in the results and discussion section, the overall average of the test scores increased. However it must be pointed out that every student who took the retest did not improve their score. The majority of students did improve their scores and they also feel positive about using computers but it is difficult to state the remediation activities are responsible for the improvement. One student felt their improvement was due to discussing the topics and questions with other students. Still the information generated by this project shows that computers can be an effective aide and that students enjoy the computer based activities created by their teacher. This alone is enough to continue to explore alternative methods of instruction and remediation with computers.

References

Anderson, J. R., Corbett, A. T., Koedinger, K. R., & Pelletier, R. (1995). Cognitive tutors: Lessons learned. Journal of the Learning Sciences, 4(2), 167.

Carswell, L., Thomas, P., & Petre, M. (2000). Distance education via the internet: The student experience. British Journal of Educational Technology, 31(1), 29-46.

Christmann, E., & Badgett, J. (1999). A comparative analysis of the effects of computer-assisted instruction on student achievement in differing science and demographical areas. The Journal of Computers in Mathematics and Science Teaching, 18(2), 135-143.

Churach, D., & Fisher, D. L. (2001). Science students surf the web: Effects on constructivist classroom environments. The Journal of Computers in Mathematics and Science Teaching, 20(2), 221-247.

Cox-Petersen, A. M., & Olson, J. K. (2000). Authentic science learning in the digital age. Learning and Leading with Technology, 27(6), 32-35, 61.

Crowther, M. S., Keller, C. C., & Waddoups, G. L. (2004). Improving the quality and effectiveness of computer-mediated instruction through usability evaluations. British Journal of Educational Technology, 35(3), 289-303.

Gibson, S. D. (1997). A comparative study of previous achievement indices for two groups of ninth-grade students: Those who passed and those who failed sections of the ohio ninth grade proficiency test. American Secondary Education, 26, 2-6.

Horwitz, P. (1999). Hypermodels: Embedding curriculum and assessment in computer-based manipulatives. Journal of Education, 181(2), 1.

Jacob, B. A. (2001). Getting tough? The impact of high school graduation exams. Educational Evaluation & Policy Analysis, 23(2), 99-121.

Kubow, P. K., & DeBard, R. (2000). Teacher perceptions of proficiency testing: A winning ohio suburban school district expresses itself. American Secondary Education, 29(2), 16-25.

Littlejohn, A., Suckling, C., & Campbell, L. (2002). The amazingly patient tutor: Students' interactions with an online carbohydrate chemistry course. British Journal of Educational Technology, 33(3), 313-321.

MacKinnon, G. R. (2001). A promising model for incorporating the computer in science learning. The Journal of Computers in Mathematics and Science Teaching, 20(4), 395-404.

Mason, D., & Crawley, F. E. (1993). Remediation, bridging explanations, worked examples and discussion: Their effectiveness as teaching strategies in a freshman-level nonscience major chemistry course. Access eric: Fulltext (143 Reports--Research; 150 Speeches/Meeting Papers). Texas.

Morse, R. H. (1991). Computer uses in secondary science education. Eric digest.

Nakhleh, M. B., Donovan, W. J., & Parrill, A. L. (2000). Evaluation of interactive technologies for chemistry websites: Educational materials for organic chemistry web site (emoc). The Journal of Computers in Mathematics and Science Teaching, 19(4), 355-378.

Neal, J. C., & Kelly, P. R. (2002). Delivering the promise of academic success through late intervention. Reading & Wrting Quarterly, 18, 101-117.

Neugent, L. W. (2004). Getting ready for online testing. T.H.E. Journal, 31(12), 34, 36.

Reising, R. W. (2000). High school exit exams. The Clearing House, 74(1), 4-5.

Rodrigues, S. (2000). The interpretive zone between software designers and a science educator: Grounding instructional multimedia design in learning theory. Journal of Research on Computing in Education, 33(1), 1-15.

Schneider, J. (2000a). Science explorer 3.04. T.H.E. Journal, 27(9), 78.

Schneider, J. (2000b). Science gateways. T.H.E. Journal, 27(9), 78.

Scott, L. (2001a). Naming chemical compounds. Learning and Leading with Technology, 29(1), 60-61.

Scott, L. (2001b). Periodic table & trends. Learning and Leading with Technology, 29(1), 60-61.

Sherman, G. (1999). Western harnett high school, lillington, north carolina. Plato evaluation series. 12.

Starratt, R. J. (2003). Opportunity to learn and the accountability agenda. Phi Delta Kappan, 85 (4), 298-303.

Tillman, S. (2000). Britannicaschool.Com. T.H.E. Journal, 28(3), 73.

Timmons, M. (2001). Hyperchemistry on the web. School Library Journal, 47(1), 60-61.

Today, U. (2004). High school graduation tests have little tie to college, report finds. USA Today.

Vogler. (2002). The impact of high-stakes, state mandated student performance assessment on teacher's instructional practices. Education, 123(1), 39.

Windschitl, M. (2000). Supporting the development of science inquiry skills with special classes of software. Educational Technology Research and Development, 48(2), 81-95.

Woodfield, K. (2003). Getting on board with online testing. T.H.E. Journal, 30(6), 32, 34-37.

 

 

Appendix 1

Interview and Survey Questions

Whole Class Interview Questions

  • What if anything did you like about the computer based review tools?
  • What if anything did you dislike about the computer based review tools?
  • Which of the activities if any did you use more than once?
  • Do you have any suggestions for making the process more helpful?
  • What would you change for future use?

Written Survey

1. How helpful was using the computers to review for the Benchmark assessment?

Very helpful Some what helpful Not very helpful

2. How much time outside of class did you spend using the computer activities to study for the Benchmark assessment?

15 minutes or less 16-30 minutes 31-45 minutes 45-60 minutes over an hour

3. How many different activities did you use?

1 2 3 4 5 more than 5

 

Appendix 2

(pre) 6 week Benchmark Assessment score   (post) 6 week Benchmark Assessment score   Score change       Average change by group   All classes    
                    Pre   Post
Individual Classes                   28.0   56.0
48.0   80.0   32.0   6 increased   13.3   48.0   52.0
64.0   64.0   0.0           52.0   96.0
68.0   68.0   0.0   4 unchanged       56.0   52.0
68.0   60.0   -8.0           56.0   72.0
68.0   72.0   4.0   2 decreased   -6.0   56.0   44.0
68.0   68.0   0.0           56.0   68.0
68.0   68.0   0.0           64.0   60.0
72.0   88.0   16.0           64.0   52.0
76.0   88.0   12.0           68.0   72.0
76.0   84.0   8.0           68.0   76.0
76.0   84.0   8.0           72.0   60.0
88.0   84.0   -4.0           72.0   76.0
                    72.0   84.0
                    72.0   68.0
70.0   75.7   5.7           80.0   84.0
                    88.0   92.0
                    96.0   96.0
                    52.0   72.0
                    56.0   84.0
52.0   72.0   20.0   15 increased   15.5   60.0   84.0
56.0   84.0   28.0           60.0   56.0
60.0   84.0   24.0           64.0   40.0
60.0   56.0   -4.0           64.0   84.0
64.0   40.0   -24.0   3 decreased   -10.7   64.0   68.0
64.0   84.0   20.0           64.0   80.0
64.0   68.0   4.0           68.0   84.0
64.0   80.0   16.0           68.0   100.0
68.0   84.0   16.0           72.0   92.0
68.0   100.0   32.0           72.0   88.0
72.0   92.0   20.0           76.0   80.0
72.0   88.0   16.0           76.0   72.0
76.0   80.0   4.0           76.0   92.0
76.0   72.0   -4.0           76.0   80.0
76.0   92.0   16.0           84.0   92.0
76.0   80.0   4.0           88.0   92.0
84.0   92.0   8.0           48.0   80.0
88.0   92.0   4.0           64.0   64.0
                    68.0   68.0
                    68.0   60.0
68.9   80.0   11.1           68.0   72.0
                    68.0   68.0
                    68.0   68.0
28.0   56.0   28.0   11 increased   12.7   72.0   88.0
48.0   52.0   4.0           76.0   88.0
52.0   96.0   44.0   1 unchanged       76.0   84.0
56.0   52.0   -4.0           76.0   84.0
56.0   72.0   16.0   6 decreased   -8.0   88.0   84.0
56.0   44.0   -12.0                
56.0   68.0   12.0       Overall        
64.0   60.0   -4.0       Average   67.7   75.2
64.0   52.0   -12.0                
68.0   72.0   4.0                
68.0   76.0   8.0                
72.0   60.0   -12.0                
72.0   76.0   4.0                
72.0   84.0   12.0                
72.0   68.0   -4.0                
80.0   84.0   4.0                
88.0   92.0   4.0                
96.0   96.0   0.0                
                         
                         
64.9   70.0   5.1                
                         

Appendix 3

 

(pre) 6 week Benchmark Assessment score   (post) 6 week Benchmark Assessment score   Score change         All classes    
                  Pre   Post
Individual Classes                 28.0   56.0
48.0   80.0   32.0   2 increased     48.0   52.0
64.0   64.0   0.0         48.0   80.0
68.0   68.0   0.0   4 unchanged     52.0   96.0
68.0   60.0   -8.0         52.0   72.0
68.0   72.0   4.0   1 decreased     56.0   52.0
68.0   68.0   0.0         56.0   72.0
68.0   68.0   0.0         56.0   44.0
                  56.0   68.0
64.6   68.6   4.0         56.0   84.0
                  60.0   84.0
52.0   72.0   20.0         60.0   56.0
56.0   84.0   28.0         64.0   60.0
60.0   84.0   24.0   8 increased     64.0   52.0
60.0   56.0   -4.0         64.0   40.0
64.0   40.0   -24.0         64.0   84.0
64.0   84.0   20.0         64.0   68.0
64.0   68.0   4.0   2 decreased     64.0   80.0
64.0   80.0   16.0         64.0   64.0
68.0   84.0   16.0         68.0   72.0
68.0   100.0   32.0         68.0   76.0
                  68.0   84.0
62.0   75.2   13.2         68.0   100.0
                  68.0   68.0
28.0   56.0   28.0         68.0   60.0
48.0   52.0   4.0   7 increased     68.0   72.0
52.0   96.0   44.0         68.0   68.0
56.0   52.0   -4.0   4 decreased     68.0   68.0
56.0   72.0   16.0              
56.0   44.0   -12.0   Overall          
56.0   68.0   12.0   Average     60.3   69.0
64.0   60.0   -4.0              
64.0   52.0   -12.0              
68.0   72.0   4.0              
68.0   76.0   8.0              
                       
56.0   63.6   7.6