I’m going to have to disagree with this statement. I, like many students, am not a good test taker. I get nervous and sometimes forget things. Students can be having a bad day, not feel well; so how can we objectively use their data?
When students need the information they are being tested on, is there a time when they will be unable to ask someone for help? Probably not. So, why do we expect them to be able to regurgitate information just so they can pass a test and we can have our data? It’s unrealistic.
I think it is okay to use data to make decisions for school improvement, but it shouldn’t be the only thing looked at. There are other factors that should be considered when making those decisions, such as how they perform in class, and what teachers (and others) believe the needs for school improvement are. I’m not saying that you need to leave the decision up to the teachers, but they are your driving force, and they need to have a voice somewhere in the process.
I am going to agree with Jill on this one. I'm not a good test taker either. I get extremely nervous and sick on the stomach knowing that my academic success or whether I get a certificate to teach relies heavily on one assessment. However, on assignments such as essays and other in class assessments, I do well. It is this type of student that has to repeat courses over and over because they didn’t pass the EOC despite the fact that they passed the class. So what is that saying? Is it saying that because a student doesn’t do well on the test, they don’t know the material, or is it saying that they just don’t test well? It can be assumed that test are only good for measuring data of students who knows test taking strategies, knows the information, and who knows how to remain calm under pressure. For these students, tests can be a good way to measure what they know or do or do not know, assuming that only what was taught in class will appear on the test. Otherwise, if teachers are not teaching exactly what is on the test, it won’t matter if students test well or not because they will all be at risk of failing anyway. Lastly, you have to take into consideration the students who guess on tests who get lucky and pass. These same students who guessed and passed could have a zero in the actual class. So that is not providing an accurate assessment of the student’s knowledge. Now if we want to see things as black or white or see things from the perspective of a politician, I will have to agree that numbers and test data do not lie, after all, what else do they have o go on? However, if you are looking at this from an educator’s perspective, then you have to realize this topic (question) is not just black and white but rather grey. If you deal in the grey area then you have to admit that tests do not accurately measure student achievement.
Data is just ONE component of what should be used to assess student achievement; and, yes it does lie. I agree with the previous three comments. There is value in collecting data when making instructional decisons, as Keith said, but just as we understand there are several different learning styles, why is it that we do not give students a choice of different assessments when it comes to EOCs and VoCATs. Can we say that students also have different testing styles.
I think that there are many strengths and weaknesses to using data for school improvement. I think that data should be a piece of information used in making these decisions. However, data should not be the only piece of information used.
I agree with many of the existing posts, but I also believe that data driven decisions have value. As Jill and Tiffany mentioned, tests do not tell the entire story of an educational environment. In some cases, learning may be best illustrated during class discussions, projects, or even when students teach the class. That said, in my experience, tests do provide important information. I don't know the entire testing landscape in NC, but I am familiar with the high school math exams. For most exams, students must answer just 50% of the questions to be declared "proficient." The typical curved grade for a 50% is a 77%. This bar is extremely low. In math at least, we can get a pretty solid indication of proficiency levels. If anything, the levels may be too generous. In closing, data does not tell the entire story...but, attendance, suspension, testing, and professional development data can help pinpoint where schools should focus school reform efforts.
I completely agree, data does have value but other measures should be used in addition to the data. I did not have trouble with the voicethread. You have to click on the link and then register with your email address before you can post.
I’ve always listened to administrators talk about how numbers don’t lie and how many decisions are usually made with numbers collected from data. I agree with Debra in that there are many other components to making decisions regarding school improvement other than using data. Although I do agree that data is an EXTREMELY important element in developing a school improvement plan, as I’ve learned and will continue to always reference, the unique cultures of schools will play an even bigger part when it comes to developing a model for school improvement.
I agree with what I have read here. Data (test scores) tells a piece of the story, and should be included in school improvement along with attendance issues and how to deal with them, professional development for teachers that deals with RTI, relationship building, student retention, professional learning communities, and effective teaching strategies. A teacher once said to me that if we were honest with ourselves, and we really looked at the students we were responsible for teaching honestly, we could accurately predict what their EOG scores would be. We don't do this, he said, because if we predict they are not going to make proficiency in our subject, we're admitting that while we might have "taught" a concept, our students didn't learn it, so what should we have done differently? We tell students to take responsibility for their learning, but are we doing the same thing for our teaching? Interestingly enough, as I've been looking over Carrington's sixth graders' CogAT scores, I find that their current ability scores are matching what their 5th grade EOG (achievement) scores show. So, long story short, you can't ignore data, but you can't let it be the whole ball of wax either. Camille
Call me crazy, but this question reminds me of the whole idea of using 3rd grade reading scores to predict the prison population. Does anyone know if this is this an urban legend or is there actual data to back up the correlation?
Many of the comments focus on the fact that a student's standardized test score doesn't paint the full picture of learning. I agree, but that's a teacher perspective. Teachers see their students daily, they have first-hand experience with their students, so they can quibble about standardized test scores. But principals have little first-hand experience with the learning of individual students.
If data (particularly standardized test score data) don't tell the whole story, then what other kinds of information should a principal be using to drive school improvement?
In response to Dr. G's question, I thought I would do a google search and I found this article. I thought it was interesting and wanted to share with the group.
I agree with everyone else. I certainly do not think numbers are so objective and we all know that we can make numbers say what we want them to say often. There are too many other factors to take into consideration when making decisions for school improvement.
Principals can use other things to gauge school success and what needs to be deal with in regards to school improvement such as teacher working conditions surveys, discussions with staff and students, observations from doing class and building walk-throughs. The other things Debra mentioned are great as well.
While test scores (strictly proficiency levels) do not tell the whole story, it seems that principals must have conversations with their teachers about growth rates. If one teacher consistently sees high individual growth rates, and another sees consistent declines, questions must be asked. Of course, these questions may be answered by other figures such as attendance rates. Then, it becomes a school-wide issue, as opposed to a classroom issue.
I agree with opening comment that test scores do not lie. The more important question is the cause or reasons for the low test scores. Two years ago, the DPS World History Benchmark did not correlate with the pacing guide so the test did not cover what was actually taught. The results of the tests were similar to the previous year's and subsequent year's tests. The determining factor in proficiency was directly related to the student's lexile scores for reading ability and comprehension. Those students who were above grade level in reading were proficient and those who were below reading level were not proficient in all three years. So the test did not actually teach the content of World History, but the student's reading level. That is the biggest problem with standardized tests in the Humanities and in general, they don't test what they are intended to test.
Dr. Graham, I think I hinted on the question you asked in my first response. I said that test scores are only useful to principals and other school stakeholders who do not teach or have exposure to what goes on inside a class. However, test scores are not the only way to monitor student achievement. Teachers could create a portfolio on each students that consist of students’ work that tracks their progress/ lack of progress throughout the year that way principals will be able to pull students folders and monitor their progress. Or, principals can actually monitor classes more frequently to observe students in class to see if the students are grasping the concepts and see that the teacher is teaching the concepts effectively. These are far from perfect solutions, but they are possible alternate ways to monitor student achievement.
I read a great article today about intervention for students when they are not considered proficient, and one statement said that teachers blame everybody and everything else...parents, administrators, politicians, lazy students, TV...what would happen if teachers looked closely at what they are doing each day in their classes that is causing students not be proficient? So data doesn't lie, but sometimes it points a finger at us(teachers). I like Tiffany's suggestion. Use test scores, portfolios and teacher evaluations to help monitor what is going on in the classroom. Some principals (like mine) only allow AIG licensed teachers teach our AIG reading and math classes. Speaking of AIG kids, they normally make proficiency, but they as a group don't make much growth. Why is that? It's not because they are at the top and have no where to go. What do teachers do to meet the needs of gifted kids? Data tells us a lot. If we use it, student portfolios, teacher performance, and student attendance and discipline records to establish our school improvement initiatives and if we truly reflect on what we are doing in our classrooms, we will see a difference in what data is saying about our schools and classrooms.
Paul makes an interesting point about test scores. The test scores themselves may not lie, but our interpretations of what they mean can be inaccurate. A history test can really be a reading comprehension test masquerading in disguise.
How can principals trust and use data if they are not sure what the data really mean?
To the original comment that the data "just don't lie"...I say this is true, but with a major league caveat. The data can paint just about any picture we want it to (sometimes).
I teach 4th grade. Believe it or not (and I say this with tongue firmly planted in cheek) some kids do not try their best on standardized tests, no matter how much I want them to. Therefore, the 1 or 2 they get may not be an accurate reflection of their ability.
On the flipside, I have had some kids that are wonderful math students when it comes to computations and try their absolute best on the EOG...but their reading comprehension is so low, they don't recognize which operation to do on the math test, and bomb it as a result. Meanwhile, they GUESS their way to a 3 on the reading part.
So...while the data may not lie in regards to whether a student passes or fails on that particular test, that same data doesn't necessarily tell the whole story either.
"How can principals trust and use data if they are not sure what the data really mean?"
To this question I would say you cannot put 100% faith in test score data solely. Even if you know EXACTLY what the data mean, one test cannot possibly paint the full picture of a student's ability.
This is where principals need to be surrounded by competent staff that can accurately, and without bias, assess the daily, weekly, quarterly progress of a student and provide their own supporting data.
I know...more data. Well...if one set of data says one thing, sometimes you have to present a set of data that tells a different story.
It's like a Hall of Fame baseball player with a lifetime .330 batting average. No one ever says "Well, he failed as a hitter 67% of the time." You focus on the positive instead.
There...I've brought baseball into the discussion, and Matt, I did so without mentioning how my beloved Yankees just one their 27th World Series title. Please forgive my self-indulgence everyone. See you Wednesday.
I also agree with everyone else. Test scores should not be the only thing used for data driven school improvement. In response to Dr. G's question, it would be helpful if principals could have data on all students from the beginning of the school year to the end of the school year. I have seen a data wall created for an entire student body of a school. It helped with tracking growth. It helped with setting up remediation. The staff knew who was getting EC services including AIG. It made everyone accountable. Everyone knew what targeted goals were needed to make growth.
Like Tom said, surrounding yourself with people who are competent who can track the progress of students is a good way to judge the accuracy of the data. Having teachers give their input on the test is another way to judge whether or not to completely trust the data. As students take the tests, teachers walk around and read questions and students answers, so they should be able to provide an insight on whether or not the test was accurately obtaining information for data to be used.
I think, too, regardless of whether the data is accurate and can be trusted, it’s what is produced at your school and you need to do something with it. We may not think that a test was objective, but there’s not any way that we can really ignore the data just because we don’t agree with it. It doesn’t matter if we think that the data may lie or not, that is still how people are going to look at the performance of our school.
I agree with much of what has been said: data is a piece to the puzzle when trying to make informed decisions. That said, I feel as if data, at least in my classroom, is a HUGE piece of the puzzle. While there are times that extenuating circumstances (such as a bad test taker) may skew the data, I think that most often times, the data tells the right story. I have tracked student data in my classroom for years now, and have found data-driven instruction to be extremely effective in my classroom. When it comes to school improvement planning, I again think that the numbers should be one of the biggest pieces of the puzzle. While again, there may be a few extenuating circumstances, I do not think that those circumstances are enough to skew the numbers. I agree that the numbers just don't lie.
The biggest thing that prinicipals need to do with the data is ask why the scores are low and figure out what needs to change to help studemts improve. As we learned in system analysis in our last course, their is no blame only problems that need to be addressed. Also, there is very rarely only area that is causing the problem or deficiency.
As a scientist, I think some people are getting confused and misled by the word data. The scientific process is all about the reflection and analysis of the data. Data itself does not lie, but it also does not tell the truth. The truth comes in the reporter of that data and how it was analyzed and compared to other data, norms and the expected data results. Therefore, anyone can report data and tell the truth but someone else can report and take the same data and get a pretty different result which is also true.
I believe that data is a critical component to any type of study you do. In order for there to be any validity to the reasoning behind the decisions you make, data is needed.
I also agree with Benji in that, if you are going to use data, you have to be mindful of the accuracy of the data, and also aware of who is conducting the research, reporting, and handling any other type of analysis that may go along with the data.
Benjy's comment reinforces my comment in response to Dr. G's second question. The principal like a scientist needs to analylize the data from sevearl sources and form conclusions as to what caused the particular outcomes and what needs to done to either replicate desired results and what needs to be change to eliminate undesirable outcomes.
Principals cannot trust data if they don't know what it means. They can rely on other's explanation of the data, which they often do. That's why it's good to use many different types of data as discussed thoroughly above.
"If data (particularly standardized test score data) don't tell the whole story, then what other kinds of information should a principal be using to drive school improvement?"
I believe that data should be the driving force behind the school improvement process. But, as others have said, there are other important pieces to the puzzle. Things such as student attendance, suspension rates, portfolios, teacher observations, and growth rates should be considered in the school improvement process.
I think that it is a principal's job to act as investigator and figure out what the scores are telling him/her. For example, whenever we take benchmarks at our school my principal calls me in for a conversation. Typically he will speak with me before the test and ask how I THINK my students will do. After we receive our scores he ALWAYS sits down with me to have a conversation about how the students did and what I believe contributing factors were. During this conversation he also puts a lot of emphasis on the growth rate.
The data doesn't mean anything until the principal interprets it. I believe that during this interpretation process anecdotal evidence should be combined with the other pieces of evidence previously mentioned (portfolios, teacher observations, etc.) to come to a conclusion about the story the data is telling.
I am a die hard cubs fan and I say next year is our year! HAHA, but anyway, the test scores and the data does give you information and I agree that the principal should be the best analyzer of that data and how that data plays a part as to how the students are doing and how the school is performing as a whole. My big problem with giving data to the public is that a school may make AYP, get a bonus and the accolades that go with it because they grew from 40% proficient to 50% proficient. While, another school in the same district gets no recognition from the state or the media yet their students are at a 70% proficiency level. Again the data is all about the analysis and how it is reported.
Below is a list of statistical truths called Ashley-Perry's statistical axioms. 1) Numbers are tools, not rules. 2) Numbers are symbols for things; the number and the thing are not the same. 3) Skill in manipulating numbers is a talent, not evidence of divine guidance. 4) Like other occult techniques of divination, the statistical method has a private jargon deliberately contrived to obscure its methods from nonpractitioners. 5) The product of an arithmetical computation is the answer to an equation; it is not the solution to a problem. 6) Arithmetical proofs of theorems that do not have arithmetical bases prove nothing.
I feel as if each one of these could be dissected and used to discuss education. I think the biggest problem is in the original statement. If the powers that be use and abuse numbers then they no longer have to actually make tough decisions. They just manipulate the data to say what they want it to say.
I think the role of a principal should be the instructional leader in the school. It should be the principal’s responsibility to understand what the data says. (Actual scores) However, if the principal needs interpretation of what the data means from other sources that should be accepted. Principals are the instructional leaders that make decisions based on data. A principal should be able to use his/her faculty as well as other district administrators for their expertise on how to interpret data. I agree with Abbey. The data doesn't mean anything until the principal interprets it. I believe that principals use the interpretations to make the best decisions possible for school improvement.
I agree that data does not lie, but people do. Data can be manipulated to support the motives of people. We have to question the source in which the data is coming from and also the context of the data. A prime example would be looking at posted EOG results. A class’s scores are only at a snapshot of one assessment. Those score alone do not show the growth or regression of the student. Data does not lie and should be used as an initial tool to single out areas to analyze closer to discover why something is effective or ineffective.
Benjy, thanks for sharing the list of statistical truths called Ashley-Perry's statistical axioms. 1) Numbers are tools, not rules. 5) The product of an arithmetical computation is the answer to an equation; it is not the solution to a problem. Numbers 1 & 5 brought a lot of thought to mind about how scores are used to drive instruction. Scores are used to drive instruction in retaining students, in promoting students and qualifying students for EC services. Numbers are tools since they are used for decision making in school improvement. The interpretation of numbers is key to the success of schools!
Schools have to be data driven. My definition of data is any thing that is used for reasoning, discussion, and/or calculation. That encompasses every thing from EOG scores to student attendance. Portfolios, student suspensions, and teacher observations are all forms of data. To often when we hear data driven in a school setting we automatically think of standardized test scores, may be because of the emphasis put on them by ourselves and others.
It is extremely important for the principal to be surrounded by people that have different strengths than them. This would aid in offsetting some of their weaknesses. If a principal is not that knowledgeable at interpreting any data he/she could misinterpret the data and mismanage a situation.
Again data does not lie but two people can look at the same exact data and come up with two totally different interpretations. The interpretation of data drives decision making. That makes me think of the saying, “Is the glass half full or half empty?”
I would disagree with this statement whole heartedly. I say that the greatest flaw with making data driven decisions is relying solely on data and not factoring in the intangibles. When working with and relying on data it is important to know how it was collected and by whom. I will agree data, more specifically, numbers don’t lie but depending on you which ever side you are on, you can find the data to support your claims.
Sometimes I look at this argument about data never lying and being the bottom-line in making important decisions in education the same way some people look at life in black or white, good or bad. Neither life nor school improvement could ever be that simple. There is something to be said for accounting for the intangibles.
In response to Dr. Graham’s question, I think that principal’s should allow teachers to discuss the different challenges and successes they feel their students are having with the expectation that help or ideas would be provided or imitated. It can be really difficult to utilize this idea at all public schools, but when principals attempt to listen to their faculty about student achievement, I sure the data collected might reveal a more accurate account as to how might to emulate or remediate the problem(s).
“Don't be buffaloed by experts and elites. Experts often possess more data than judgment. Elites can become so inbred that they produce haemophiliacs who bleed to death as soon as they are nicked by the real world.”
I would have to say that I disagree with the initial statement. Although we think of data as objective, we must consider the context in which they are used. I think this is especiall the case when we think about measures such as norm-based tests. Upon who was the norm based? Thinking back to Dr. English's class, we learned that for many of these norm based tests such as SATs and such, the norm often represented an atypical population that was not very represenatitive of the tested population. Therefore, I think we must be careful to consider contexts in which data is produced before adding value to it.
I think another thing that we must consider when collecting data is how thorough we are in the process of the data collection. I know that in terms of surveys (the kinds I can think of for School Improvement are those relating to parent satisifaction, teacher satisfaction, etc...), response rate can play a big role in deciding if the data can be trusted. I can think of a particularly valid example from this past school year. As a school, we took a survey developed by the goal manager of the teacher recruitment and retention goal regarding staff attitudes and ways the school could improve. Many staff responded with comments that were not favorable with the status quo of the way things had been going. The following staff meeting, our Principal went on a tirade about how if she had a magic wand, she would wave it, and that she would like to know who thinks that the administration isn't doing xy or z...you get the point. Anyways, when the next survey rolled around, many staff members did not participate because the previous time they did, their ideas were belittled. Well, lo and behold scores on the measures did improve. However, fewer people responded. What our administrator did was silenced those who had opinions that were not supportive of her work. Those that did choose to respond to the second survey were not as candid in their feedback. Although this is a long story, I think it well illustrates the point that we must consider circumstances surrounding data collection, for instance, response rate, when thinking about making decisions with data.
I think Lamont brought up a good point about school executives surrounding themselves with a variety of different perspectives. As a school executive, we will want to have people around us who are data-driven as well as those who focus more on factors that can not be easily measured with data. As an administrator, we will also want to be skilled in looking at different kinds of data--both quantative data as well as qualatitive data. We can not be overly reliant upon just one type of data to measure holistically success or failure. To do so is to basically set ourselves up for failure when the data does not turn out the way we want it. We must learn how to read data and use it to create interpretations that are beneficial to particular situations, individuals, and decisions.
I think the following comments by Justin are very valid. I also believe that principals need to base a great deal of their decision making not only on data but also on personal experience and knowledge of the building they are in. I totally agree with the quote by Ryan specifically the piece which states "Experts often possess more data than judgment.?" Although I am for data based decision making on a lot of cases, I also believe that sometime you just have to go with you gut and experience.
An interesting comment I heard while being on campus this past week in the School of Education was a former professor mentioning that EOGs and EOCs are ultimately, minimal skills tests. So, the majority of students should be able to reason and think through questions to arrive at correct answers. When I think of this assertion and think of the test, I tend to agree. But, when I reflect on what I am doing or not doing, as Camille mentioned above, I realize the challenge involved in helping students to pass. When referring to helping students to pass, I am referring to students who are do not have an EC or ESL label. (I CERTAINLY believe that that ESL and EC students can and should be expected to achieve at high levels, but when considering these tests, I want to critically consider the issues without any unpredictable variables.)
So, if students "should" be able to think and reason through these tests, at least by and large, we have to examine why they are not being successful. As Camille pointed out, we have deficiencies in teaching. As Paul pointed out, reading is the necessary key to even have a chance at doing your best on a test. It is always interesting to me, that teachers spend more time making excuses for test scores as opposed to highlighting the need for pedagogical improvements and reading interventions that extend beyond elementary school.
In referencing Dr. G's comment about what should an administrator look to if EOG / EOC scores are the only data around, I am wondering the part that PLCs play in improving this challenge. If PLCs (or teams, or grade levels) have regular common assessments, tied to specific, necessary objectives, hopefully utilizing more than just multiple choice testing data, then this should provide ongoing, and hopefully more authentic data for teachers and administrators to discuss together. If PLCs are effectively meeting to regularly discuss students and ensure that remediation or alternate assessments are being used, then teachers and administrators will have more useful data to work with.
The ultimate challenge, from my perspective, comes when everyone "agrees" that yes, with this score, a student fails a test. The problem from my perspective is worsened, when everyone assumes, without a specific discussion that either, the student cannot read at a high enough level, the student probably just had a bad day, the student does not take tests well, the student did not try, etc. These generalizations to explain a failing score rarely address the heart of the matter, and do not hold any person / PLC accountable to providing what is necessary for the student to improve.
I think it is vital for discussions to take place for students not achieving at the expected level. When students are not at that level, then a team of professionals should be responsible, in my opinion, for helping that student either master primary literacy skills, have a vested interest in doing well, or to receive reteaching in an area of weakness.
Several postings have also mentioned that data doesn’t lie but the interpretation of the data can be manipulated to say what you want it to say. I agree that a principal, as Justin stated, should look at different types of data in order to make an informed decision about school improvement.
I really do appreciate the statistical axioms that Benjy shared because the numbers are just tools, manipulating the numbers is a talent, and the product of an arithmetical computation is the answer to an equation; it is not the solution to a problem. What data is collected and how it is collected is just as important as to how the data is analyzed.
Rick, I disagree with this statement because data and statistics are ultimately a form of information that can be altered or manipulated to express a point. Not that I am a conspiracy theorist, but I was told by an educated man that there are three types of information: the truth, lies, and statistics. Regardless if you are stating the truth or a lie about an issue, generally data will be presented to support that view or statement. Simply put, data is an ambigious venue/platform to create persuasion.
Conversely, data-driven decisions are justified at times. Like anything else, data and data-driven decision should be taken and made in moderation. Indicating that data is objective is an ambiguous staement, that is leaning toward false, as data is presented personify a particular area that needs to be address. Case and point, a school overall may have perform well on the EOGs last year, but someone wants to focus on the EC students'scores for the 7th grade. That person will be their crusade by highlighting the scores of all EC students, then in each grade. Depending on the individual's agenda, he/she could advocate for better resources or attempt to dismantle an EC team or teacher. So once again, data-driven decisions should be done over careful consideration and in moderation.
As I consider the advantages and disadvantages of standardized testing, I realize that my opinion shifts as I think from the perspective of a teacher, an administrator, or a person who is generally concerned regarding the current state of public education. As a teacher, it is disheartening and sometimes insulting when people judge my effectiveness solely on testing data, without considering the composition of my class and especially the individual growth that usually accompanies a failing score. As a principal, there seems to be little data that is comparable with other schools and objective as some common assessments and most portfolios are generally subjective. As an average person considering the state of public schools, standardized testing or not, it is hard to ignore that some subgroups are failed in public education. The more and more I think of the advantages and disadvantages of standardized testing, it does to seem to serve a specific purpose. If only standardized testing could complement a more authentic way of evaluating students' learning, then I think we would have made progress to fairly comparing the successes of all students while allowing different assessments to showcase student learning in ways appropriate to the individual student.
I strongly agree with the notion that the data does not lie. Therefore, the strength of a data driven decision-making process is collecting and analyzing information accurately. My belief is because an organization cannot make improvements until they have a sense of their baseline data.
If data (particularly standardized test score data) don't tell the whole story, then what other kinds of information should a principal be using to drive school improvement? Principals should collaborate with department chairs, grade level chairs, and the classroom teacher to analyze at least three forms of data that have a direct impact on student achievement: (1) daily school attendance, (2) cumulative assessment data, and (3) discipline referrals. Daily school attendance: If students do not attend school on a consistent basis, he or she can learn the information taught by the classroom teacher. Therefore, principals need to assemble a team that will be responsible for ensure that students that show signs of difficulty with attending school regularly receive support from the school to increase the likelihood that the student will attend. One strategy may be as simple as having a caring educator call the student’s home every morning. Again, a large part of the success in education is based on relationships, not how intelligent the teacher may be. Cumulative assessments: Teachers should record the instructional delivery methods used to teach objectives in their Lesson Plan Book. Based upon the student achievement results on classroom discussions, quizzes, tests, and projects; the principal should be able to assess which deliveries methods were effective. Additionally, the same data should be used to determine which students struggled with specific objectives. This information should be used to remediate struggling students. Discipline referrals: This data will give the principal a sense of how effective the teacher’s classroom management system is working. If the teacher has a firm grip on classroom management, it generally decreases the number of distraction from opportunities to deliver classroom instruction.
How can principals trust and use data if they are not sure what the data really mean? Because principals should model continuous learning, I would recommend that he or she contact the following district offices for support: (1) Testing – testing data (2) Curriculum and Instruction – instructional content and delivery methods, and (3) Prevention Services – interventions for at-risk students (attendance or behavioral problems). Once the principal gains a deeper understanding, they should allow the district’s experts to engage the entire staff in discussions that will lead to a greater understanding of the data collected. Finally, the staff would collaborate by pinpointing appropriate intervention strategies for struggling students.
I’m going to have to disagree with this statement. I, like many students, am not a good test taker. I get nervous and sometimes forget things. Students can be having a bad day, not feel well; so how can we objectively use their data?
ReplyDeleteWhen students need the information they are being tested on, is there a time when they will be unable to ask someone for help? Probably not. So, why do we expect them to be able to regurgitate information just so they can pass a test and we can have our data? It’s unrealistic.
I think it is okay to use data to make decisions for school improvement, but it shouldn’t be the only thing looked at. There are other factors that should be considered when making those decisions, such as how they perform in class, and what teachers (and others) believe the needs for school improvement are. I’m not saying that you need to leave the decision up to the teachers, but they are your driving force, and they need to have a voice somewhere in the process.
I am going to agree with Jill on this one. I'm not a good test taker either. I get extremely nervous and sick on the stomach knowing that my academic success or whether I get a certificate to teach relies heavily on one assessment. However, on assignments such as essays and other in class assessments, I do well. It is this type of student that has to repeat courses over and over because they didn’t pass the EOC despite the fact that they passed the class. So what is that saying? Is it saying that because a student doesn’t do well on the test, they don’t know the material, or is it saying that they just don’t test well? It can be assumed that test are only good for measuring data of students who knows test taking strategies, knows the information, and who knows how to remain calm under pressure. For these students, tests can be a good way to measure what they know or do or do not know, assuming that only what was taught in class will appear on the test. Otherwise, if teachers are not teaching exactly what is on the test, it won’t matter if students test well or not because they will all be at risk of failing anyway. Lastly, you have to take into consideration the students who guess on tests who get lucky and pass. These same students who guessed and passed could have a zero in the actual class. So that is not providing an accurate assessment of the student’s knowledge. Now if we want to see things as black or white or see things from the perspective of a politician, I will have to agree that numbers and test data do not lie, after all, what else do they have o go on? However, if you are looking at this from an educator’s perspective, then you have to realize this topic (question) is not just black and white but rather grey. If you deal in the grey area then you have to admit that tests do not accurately measure student achievement.
ReplyDeleteI disagree with the statement, but I believe there is value in collecting data that is used in making instructional decisions.
ReplyDeleteData is just ONE component of what should be used to assess student achievement; and, yes it does lie. I agree with the previous three comments. There is value in collecting data when making instructional decisons, as Keith said, but just as we understand there are several different learning styles, why is it that we do not give students a choice of different assessments when it comes to EOCs and VoCATs. Can we say that students also have different testing styles.
ReplyDeleteI think that there are many strengths and weaknesses to using data for school improvement. I think that data should be a piece of information used in making these decisions. However, data should not be the only piece of information used.
ReplyDeleteI agree with many of the existing posts, but I also believe that data driven decisions have value. As Jill and Tiffany mentioned, tests do not tell the entire story of an educational environment. In some cases, learning may be best illustrated during class discussions, projects, or even when students teach the class.
ReplyDeleteThat said, in my experience, tests do provide important information. I don't know the entire testing landscape in NC, but I am familiar with the high school math exams. For most exams, students must answer just 50% of the questions to be declared "proficient." The typical curved grade for a 50% is a 77%. This bar is extremely low. In math at least, we can get a pretty solid indication of proficiency levels. If anything, the levels may be too generous.
In closing, data does not tell the entire story...but, attendance, suspension, testing, and professional development data can help pinpoint where schools should focus school reform efforts.
Is anyone else having a tough time getting to the voicethread?
ReplyDeleteMatt,
ReplyDeleteI completely agree, data does have value but other measures should be used in addition to the data. I did not have trouble with the voicethread. You have to click on the link and then register with your email address before you can post.
Amatullah - Thanks for the response. It must just be my home computer. Thanks again!
ReplyDeleteI’ve always listened to administrators talk about how numbers don’t lie and how many decisions are usually made with numbers collected from data. I agree with Debra in that there are many other components to making decisions regarding school improvement other than using data. Although I do agree that data is an EXTREMELY important element in developing a school improvement plan, as I’ve learned and will continue to always reference, the unique cultures of schools will play an even bigger part when it comes to developing a model for school improvement.
ReplyDeleteI agree with what I have read here. Data (test scores) tells a piece of the story, and should be included in school improvement along with attendance issues and how to deal with them, professional development for teachers that deals with RTI, relationship building, student retention, professional learning communities, and effective teaching strategies. A teacher once said to me that if we were honest with ourselves, and we really looked at the students we were responsible for teaching honestly, we could accurately predict what their EOG scores would be. We don't do this, he said, because if we predict they are not going to make proficiency in our subject, we're admitting that while we might have "taught" a concept, our students didn't learn it, so what should we have done differently? We tell students to take responsibility for their learning, but are we doing the same thing for our teaching? Interestingly enough, as I've been looking over Carrington's sixth graders' CogAT scores, I find that their current ability scores are matching what their 5th grade EOG (achievement) scores show. So, long story short, you can't ignore data, but you can't let it be the whole ball of wax either. Camille
ReplyDeleteCall me crazy, but this question reminds me of the whole idea of using 3rd grade reading scores to predict the prison population. Does anyone know if this is this an urban legend or is there actual data to back up the correlation?
ReplyDeleteMany of the comments focus on the fact that a student's standardized test score doesn't paint the full picture of learning. I agree, but that's a teacher perspective. Teachers see their students daily, they have first-hand experience with their students, so they can quibble about standardized test scores. But principals have little first-hand experience with the learning of individual students.
ReplyDeleteIf data (particularly standardized test score data) don't tell the whole story, then what other kinds of information should a principal be using to drive school improvement?
Dr. G
I think other information such as attendance, suspensions, nutrition and other achievement test data (ex. lexiles) would be useful.
ReplyDeleteIn response to Dr. G's question, I thought I would do a google search and I found this article. I thought it was interesting and wanted to share with the group.
ReplyDeletehttp://www.ericdigests.org/1997-3/data.html
Here is an older article from NCDPI about using data for School Improvment.
ReplyDeletehttp://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/16/04/6d.pdf
Hope you enjoy this one also. I just skimmed a portion of the article, mostly the "executive summary" area to understand what this article was about.
I agree with everyone else. I certainly do not think numbers are so objective and we all know that we can make numbers say what we want them to say often. There are too many other factors to take into consideration when making decisions for school improvement.
ReplyDeletePrincipals can use other things to gauge school success and what needs to be deal with in regards to school improvement such as teacher working conditions surveys, discussions with staff and students, observations from doing class and building walk-throughs. The other things Debra mentioned are great as well.
ReplyDeleteWhile test scores (strictly proficiency levels) do not tell the whole story, it seems that principals must have conversations with their teachers about growth rates. If one teacher consistently sees high individual growth rates, and another sees consistent declines, questions must be asked. Of course, these questions may be answered by other figures such as attendance rates. Then, it becomes a school-wide issue, as opposed to a classroom issue.
ReplyDeleteI agree with opening comment that test scores do not lie. The more important question is the cause or reasons for the low test scores. Two years ago, the DPS World History Benchmark did not correlate with the pacing guide so the test did not cover what was actually taught. The results of the tests were similar to the previous year's and subsequent year's tests. The determining factor in proficiency was directly related to the student's lexile scores for reading ability and comprehension. Those students who were above grade level in reading were proficient and those who were below reading level were not proficient in all three years. So the test did not actually teach the content of World History, but the student's reading level. That is the biggest problem with standardized tests in the Humanities and in general, they don't test what they are intended to test.
ReplyDeleteDr. Graham, I think I hinted on the question you asked in my first response. I said that test scores are only useful to principals and other school stakeholders who do not teach or have exposure to what goes on inside a class. However, test scores are not the only way to monitor student achievement. Teachers could create a portfolio on each students that consist of students’ work that tracks their progress/ lack of progress throughout the year that way principals will be able to pull students folders and monitor their progress. Or, principals can actually monitor classes more frequently to observe students in class to see if the students are grasping the concepts and see that the teacher is teaching the concepts effectively. These are far from perfect solutions, but they are possible alternate ways to monitor student achievement.
ReplyDeleteI read a great article today about intervention for students when they are not considered proficient, and one statement said that teachers blame everybody and everything else...parents, administrators, politicians, lazy students, TV...what would happen if teachers looked closely at what they are doing each day in their classes that is causing students not be proficient? So data doesn't lie, but sometimes it points a finger at us(teachers). I like Tiffany's suggestion. Use test scores, portfolios and teacher evaluations to help monitor what is going on in the classroom. Some principals (like mine) only allow AIG licensed teachers teach our AIG reading and math classes. Speaking of AIG kids, they normally make proficiency, but they as a group don't make much growth. Why is that? It's not because they are at the top and have no where to go. What do teachers do to meet the needs of gifted kids? Data tells us a lot. If we use it, student portfolios, teacher performance, and student attendance and discipline records to establish our school improvement initiatives and if we truly reflect on what we are doing in our classrooms, we will see a difference in what data is saying about our schools and classrooms.
ReplyDeletePaul makes an interesting point about test scores. The test scores themselves may not lie, but our interpretations of what they mean can be inaccurate. A history test can really be a reading comprehension test masquerading in disguise.
ReplyDeleteHow can principals trust and use data if they are not sure what the data really mean?
Dr. G
I think it is impossible to trust anything that you do not understand.
ReplyDeleteTo the original comment that the data "just don't lie"...I say this is true, but with a major league caveat. The data can paint just about any picture we want it to (sometimes).
ReplyDeleteI teach 4th grade. Believe it or not (and I say this with tongue firmly planted in cheek) some kids do not try their best on standardized tests, no matter how much I want them to. Therefore, the 1 or 2 they get may not be an accurate reflection of their ability.
On the flipside, I have had some kids that are wonderful math students when it comes to computations and try their absolute best on the EOG...but their reading comprehension is so low, they don't recognize which operation to do on the math test, and bomb it as a result. Meanwhile, they GUESS their way to a 3 on the reading part.
So...while the data may not lie in regards to whether a student passes or fails on that particular test, that same data doesn't necessarily tell the whole story either.
"How can principals trust and use data if they are not sure what the data really mean?"
ReplyDeleteTo this question I would say you cannot put 100% faith in test score data solely. Even if you know EXACTLY what the data mean, one test cannot possibly paint the full picture of a student's ability.
This is where principals need to be surrounded by competent staff that can accurately, and without bias, assess the daily, weekly, quarterly progress of a student and provide their own supporting data.
I know...more data. Well...if one set of data says one thing, sometimes you have to present a set of data that tells a different story.
It's like a Hall of Fame baseball player with a lifetime .330 batting average. No one ever says "Well, he failed as a hitter 67% of the time." You focus on the positive instead.
There...I've brought baseball into the discussion, and Matt, I did so without mentioning how my beloved Yankees just one their 27th World Series title. Please forgive my self-indulgence everyone. See you Wednesday.
I also agree with everyone else. Test scores should not be the only thing used for data driven school improvement. In response to Dr. G's question, it would be helpful if principals could have data on all students from the beginning of the school year to the end of the school year. I have seen a data wall created for an entire student body of a school. It helped with tracking growth. It helped with setting up remediation. The staff knew who was getting EC services including AIG. It made everyone accountable. Everyone knew what targeted goals were needed to make growth.
ReplyDeleteLike Tom said, surrounding yourself with people who are competent who can track the progress of students is a good way to judge the accuracy of the data. Having teachers give their input on the test is another way to judge whether or not to completely trust the data. As students take the tests, teachers walk around and read questions and students answers, so they should be able to provide an insight on whether or not the test was accurately obtaining information for data to be used.
ReplyDeleteI think, too, regardless of whether the data is accurate and can be trusted, it’s what is produced at your school and you need to do something with it. We may not think that a test was objective, but there’s not any way that we can really ignore the data just because we don’t agree with it. It doesn’t matter if we think that the data may lie or not, that is still how people are going to look at the performance of our school.
I agree with much of what has been said: data is a piece to the puzzle when trying to make informed decisions. That said, I feel as if data, at least in my classroom, is a HUGE piece of the puzzle. While there are times that extenuating circumstances (such as a bad test taker) may skew the data, I think that most often times, the data tells the right story. I have tracked student data in my classroom for years now, and have found data-driven instruction to be extremely effective in my classroom. When it comes to school improvement planning, I again think that the numbers should be one of the biggest pieces of the puzzle. While again, there may be a few extenuating circumstances, I do not think that those circumstances are enough to skew the numbers. I agree that the numbers just don't lie.
ReplyDeleteThe biggest thing that prinicipals need to do with the data is ask why the scores are low and figure out what needs to change to help studemts improve. As we learned in system analysis in our last course, their is no blame only problems that need to be addressed. Also, there is very rarely only area that is causing the problem or deficiency.
ReplyDeleteAs a scientist, I think some people are getting confused and misled by the word data. The scientific process is all about the reflection and analysis of the data. Data itself does not lie, but it also does not tell the truth. The truth comes in the reporter of that data and how it was analyzed and compared to other data, norms and the expected data results. Therefore, anyone can report data and tell the truth but someone else can report and take the same data and get a pretty different result which is also true.
ReplyDeleteI believe that data is a critical component to any type of study you do. In order for there to be any validity to the reasoning behind the decisions you make, data is needed.
ReplyDeleteI also agree with Benji in that, if you are going to use data, you have to be mindful of the accuracy of the data, and also aware of who is conducting the research, reporting, and handling any other type of analysis that may go along with the data.
ReplyDeleteBenjy's comment reinforces my comment in response to Dr. G's second question. The principal like a scientist needs to analylize the data from sevearl sources and form conclusions as to what caused the particular outcomes and what needs to done to either replicate desired results and what needs to be change to eliminate undesirable outcomes.
ReplyDeletePrincipals cannot trust data if they don't know what it means. They can rely on other's explanation of the data, which they often do. That's why it's good to use many different types of data as discussed thoroughly above.
ReplyDelete"If data (particularly standardized test score data) don't tell the whole story, then what other kinds of information should a principal be using to drive school improvement?"
ReplyDeleteI believe that data should be the driving force behind the school improvement process. But, as others have said, there are other important pieces to the puzzle. Things such as student attendance, suspension rates, portfolios, teacher observations, and growth rates should be considered in the school improvement process.
I think that it is a principal's job to act as investigator and figure out what the scores are telling him/her. For example, whenever we take benchmarks at our school my principal calls me in for a conversation. Typically he will speak with me before the test and ask how I THINK my students will do. After we receive our scores he ALWAYS sits down with me to have a conversation about how the students did and what I believe contributing factors were. During this conversation he also puts a lot of emphasis on the growth rate.
ReplyDeleteThe data doesn't mean anything until the principal interprets it. I believe that during this interpretation process anecdotal evidence should be combined with the other pieces of evidence previously mentioned (portfolios, teacher observations, etc.) to come to a conclusion about the story the data is telling.
I am a die hard cubs fan and I say next year is our year! HAHA, but anyway, the test scores and the data does give you information and I agree that the principal should be the best analyzer of that data and how that data plays a part as to how the students are doing and how the school is performing as a whole. My big problem with giving data to the public is that a school may make AYP, get a bonus and the accolades that go with it because they grew from 40% proficient to 50% proficient. While, another school in the same district gets no recognition from the state or the media yet their students are at a 70% proficiency level. Again the data is all about the analysis and how it is reported.
ReplyDeleteBelow is a list of statistical truths called Ashley-Perry's statistical axioms.
ReplyDelete1) Numbers are tools, not rules.
2) Numbers are symbols for things; the number and the thing are not the same.
3) Skill in manipulating numbers is a talent, not evidence of divine guidance.
4) Like other occult techniques of divination, the statistical method has a private jargon deliberately contrived to obscure its methods from nonpractitioners.
5) The product of an arithmetical computation is the answer to an equation; it is not the solution to a problem.
6) Arithmetical proofs of theorems that do not have arithmetical bases prove nothing.
I feel as if each one of these could be dissected and used to discuss education. I think the biggest problem is in the original statement. If the powers that be use and abuse numbers then they no longer have to actually make tough decisions. They just manipulate the data to say what they want it to say.
I think the role of a principal should be the instructional leader in the school. It should be the principal’s responsibility to understand what the data says. (Actual scores) However, if the principal needs interpretation of what the data means from other sources that should be accepted. Principals are the instructional leaders that make decisions based on data. A principal should be able to use his/her faculty as well as other district administrators for their expertise on how to interpret data. I agree with Abbey. The data doesn't mean anything until the principal interprets it. I believe that principals use the interpretations to make the best decisions possible for school improvement.
ReplyDeleteI agree that data does not lie, but people do. Data can be manipulated to support the motives of people. We have to question the source in which the data is coming from and also the context of the data. A prime example would be looking at posted EOG results. A class’s scores are only at a snapshot of one assessment. Those score alone do not show the growth or regression of the student. Data does not lie and should be used as an initial tool to single out areas to analyze closer to discover why something is effective or ineffective.
ReplyDeleteBenjy, thanks for sharing the list of statistical truths called Ashley-Perry's statistical axioms.
ReplyDelete1) Numbers are tools, not rules.
5) The product of an arithmetical computation is the answer to an equation; it is not the solution to a problem.
Numbers 1 & 5 brought a lot of thought to mind about how scores are used to drive instruction. Scores are used to drive instruction in retaining students, in promoting students and qualifying students for EC services. Numbers are tools since they are used for decision making in school improvement. The interpretation of numbers is key to the success of schools!
Schools have to be data driven. My definition of data is any thing that is used for reasoning, discussion, and/or calculation. That encompasses every thing from EOG scores to student attendance. Portfolios, student suspensions, and teacher observations are all forms of data. To often when we hear data driven in a school setting we automatically think of standardized test scores, may be because of the emphasis put on them by ourselves and others.
ReplyDeleteIt is extremely important for the principal to be surrounded by people that have different strengths than them. This would aid in offsetting some of their weaknesses. If a principal is not that knowledgeable at interpreting any data he/she could misinterpret the data and mismanage a situation.
ReplyDeleteAgain data does not lie but two people can look at the same exact data and come up with two totally different interpretations. The interpretation of data drives decision making. That makes me think of the saying, “Is the glass half full or half empty?”
I would disagree with this statement whole heartedly. I say that the greatest flaw with making data driven decisions is relying solely on data and not factoring in the intangibles. When working with and relying on data it is important to know how it was collected and by whom. I will agree data, more specifically, numbers don’t lie but depending on you which ever side you are on, you can find the data to support your claims.
ReplyDeleteSometimes I look at this argument about data never lying and being the bottom-line in making important decisions in education the same way some people look at life in black or white, good or bad. Neither life nor school improvement could ever be that simple. There is something to be said for accounting for the intangibles.
ReplyDeleteIn response to Dr. Graham’s question, I think that principal’s should allow teachers to discuss the different challenges and successes they feel their students are having with the expectation that help or ideas would be provided or imitated. It can be really difficult to utilize this idea at all public schools, but when principals attempt to listen to their faculty about student achievement, I sure the data collected might reveal a more accurate account as to how might to emulate or remediate the problem(s).
ReplyDelete“Don't be buffaloed by experts and elites. Experts often possess more data than judgment. Elites can become so inbred that they produce haemophiliacs who bleed to death as soon as they are nicked by the real world.”
ReplyDeleteColin Powell,
US Secretary of State
I would have to say that I disagree with the initial statement. Although we think of data as objective, we must consider the context in which they are used. I think this is especiall the case when we think about measures such as norm-based tests. Upon who was the norm based? Thinking back to Dr. English's class, we learned that for many of these norm based tests such as SATs and such, the norm often represented an atypical population that was not very represenatitive of the tested population. Therefore, I think we must be careful to consider contexts in which data is produced before adding value to it.
ReplyDeleteI think another thing that we must consider when collecting data is how thorough we are in the process of the data collection. I know that in terms of surveys (the kinds I can think of for School Improvement are those relating to parent satisifaction, teacher satisfaction, etc...), response rate can play a big role in deciding if the data can be trusted. I can think of a particularly valid example from this past school year. As a school, we took a survey developed by the goal manager of the teacher recruitment and retention goal regarding staff attitudes and ways the school could improve. Many staff responded with comments that were not favorable with the status quo of the way things had been going. The following staff meeting, our Principal went on a tirade about how if she had a magic wand, she would wave it, and that she would like to know who thinks that the administration isn't doing xy or z...you get the point. Anyways, when the next survey rolled around, many staff members did not participate because the previous time they did, their ideas were belittled. Well, lo and behold scores on the measures did improve. However, fewer people responded. What our administrator did was silenced those who had opinions that were not supportive of her work. Those that did choose to respond to the second survey were not as candid in their feedback. Although this is a long story, I think it well illustrates the point that we must consider circumstances surrounding data collection, for instance, response rate, when thinking about making decisions with data.
ReplyDeleteI think Lamont brought up a good point about school executives surrounding themselves with a variety of different perspectives. As a school executive, we will want to have people around us who are data-driven as well as those who focus more on factors that can not be easily measured with data. As an administrator, we will also want to be skilled in looking at different kinds of data--both quantative data as well as qualatitive data. We can not be overly reliant upon just one type of data to measure holistically success or failure. To do so is to basically set ourselves up for failure when the data does not turn out the way we want it. We must learn how to read data and use it to create interpretations that are beneficial to particular situations, individuals, and decisions.
ReplyDeleteI think the following comments by Justin are very valid. I also believe that principals need to base a great deal of their decision making not only on data but also on personal experience and knowledge of the building they are in. I totally agree with the quote by Ryan specifically the piece which states "Experts often possess more data than judgment.?" Although I am for data based decision making on a lot of cases, I also believe that sometime you just have to go with you gut and experience.
ReplyDeleteAn interesting comment I heard while being on campus this past week in the School of Education was a former professor mentioning that EOGs and EOCs are ultimately, minimal skills tests. So, the majority of students should be able to reason and think through questions to arrive at correct answers. When I think of this assertion and think of the test, I tend to agree. But, when I reflect on what I am doing or not doing, as Camille mentioned above, I realize the challenge involved in helping students to pass. When referring to helping students to pass, I am referring to students who are do not have an EC or ESL label. (I CERTAINLY believe that that ESL and EC students can and should be expected to achieve at high levels, but when considering these tests, I want to critically consider the issues without any unpredictable variables.)
ReplyDeleteSo, if students "should" be able to think and reason through these tests, at least by and large, we have to examine why they are not being successful. As Camille pointed out, we have deficiencies in teaching. As Paul pointed out, reading is the necessary key to even have a chance at doing your best on a test. It is always interesting to me, that teachers spend more time making excuses for test scores as opposed to highlighting the need for pedagogical improvements and reading interventions that extend beyond elementary school.
In referencing Dr. G's comment about what should an administrator look to if EOG / EOC scores are the only data around, I am wondering the part that PLCs play in improving this challenge. If PLCs (or teams, or grade levels) have regular common assessments, tied to specific, necessary objectives, hopefully utilizing more than just multiple choice testing data, then this should provide ongoing, and hopefully more authentic data for teachers and administrators to discuss together. If PLCs are effectively meeting to regularly discuss students and ensure that remediation or alternate assessments are being used, then teachers and administrators will have more useful data to work with.
ReplyDeleteThe ultimate challenge, from my perspective, comes when everyone "agrees" that yes, with this score, a student fails a test. The problem from my perspective is worsened, when everyone assumes, without a specific discussion that either, the student cannot read at a high enough level, the student probably just had a bad day, the student does not take tests well, the student did not try, etc. These generalizations to explain a failing score rarely address the heart of the matter, and do not hold any person / PLC accountable to providing what is necessary for the student to improve.
I think it is vital for discussions to take place for students not achieving at the expected level. When students are not at that level, then a team of professionals should be responsible, in my opinion, for helping that student either master primary literacy skills, have a vested interest in doing well, or to receive reteaching in an area of weakness.
Several postings have also mentioned that data doesn’t lie but the interpretation of the data can be manipulated to say what you want it to say. I agree that a principal, as Justin stated, should look at different types of data in order to make an informed decision about school improvement.
ReplyDeleteI really do appreciate the statistical axioms that Benjy shared because the numbers are just tools, manipulating the numbers is a talent, and the product of an arithmetical computation is the answer to an equation; it is not the solution to a problem. What data is collected and how it is collected is just as important as to how the data is analyzed.
Rick,
ReplyDeleteI disagree with this statement because data and statistics are ultimately a form of information that can be altered or manipulated to express a point. Not that I am a conspiracy theorist, but I was told by an educated man that there are three types of information: the truth, lies, and statistics. Regardless if you are stating the truth or a lie about an issue, generally data will be presented to support that view or statement. Simply put, data is an ambigious venue/platform to create persuasion.
Conversely, data-driven decisions are justified at times. Like anything else, data and data-driven decision should be taken and made in moderation. Indicating that data is objective is an ambiguous staement, that is leaning toward false, as data is presented personify a particular area that needs to be address. Case and point, a school overall may have perform well on the EOGs last year, but someone wants to focus on the EC students'scores for the 7th grade. That person will be their crusade by highlighting the scores of all EC students, then in each grade. Depending on the individual's agenda, he/she could advocate for better resources or attempt to dismantle an EC team or teacher. So once again, data-driven decisions should be done over careful consideration and in moderation.
ReplyDeleteAs I consider the advantages and disadvantages of standardized testing, I realize that my opinion shifts as I think from the perspective of a teacher, an administrator, or a person who is generally concerned regarding the current state of public education. As a teacher, it is disheartening and sometimes insulting when people judge my effectiveness solely on testing data, without considering the composition of my class and especially the individual growth that usually accompanies a failing score. As a principal, there seems to be little data that is comparable with other schools and objective as some common assessments and most portfolios are generally subjective. As an average person considering the state of public schools, standardized testing or not, it is hard to ignore that some subgroups are failed in public education. The more and more I think of the advantages and disadvantages of standardized testing, it does to seem to serve a specific purpose. If only standardized testing could complement a more authentic way of evaluating students' learning, then I think we would have made progress to fairly comparing the successes of all students while allowing different assessments to showcase student learning in ways appropriate to the individual student.
ReplyDeleteI strongly agree with the notion that the data does not lie. Therefore, the strength of a data driven decision-making process is collecting and analyzing information accurately. My belief is because an organization cannot make improvements until they have a sense of their baseline data.
ReplyDeleteIf data (particularly standardized test score data) don't tell the whole story, then what other kinds of information should a principal be using to drive school improvement? Principals should collaborate with department chairs, grade level chairs, and the classroom teacher to analyze at least three forms of data that have a direct impact on student achievement: (1) daily school attendance, (2) cumulative assessment data, and (3) discipline referrals.
ReplyDeleteDaily school attendance: If students do not attend school on a consistent basis, he or she can learn the information taught by the classroom teacher. Therefore, principals need to assemble a team that will be responsible for ensure that students that show signs of difficulty with attending school regularly receive support from the school to increase the likelihood that the student will attend. One strategy may be as simple as having a caring educator call the student’s home every morning. Again, a large part of the success in education is based on relationships, not how intelligent the teacher may be.
Cumulative assessments: Teachers should record the instructional delivery methods used to teach objectives in their Lesson Plan Book. Based upon the student achievement results on classroom discussions, quizzes, tests, and projects; the principal should be able to assess which deliveries methods were effective. Additionally, the same data should be used to determine which students struggled with specific objectives. This information should be used to remediate struggling students.
Discipline referrals: This data will give the principal a sense of how effective the teacher’s classroom management system is working. If the teacher has a firm grip on classroom management, it generally decreases the number of distraction from opportunities to deliver classroom instruction.
How can principals trust and use data if they are not sure what the data really mean?
ReplyDeleteBecause principals should model continuous learning, I would recommend that he or she contact the following district offices for support: (1) Testing – testing data (2) Curriculum and Instruction – instructional content and delivery methods, and (3) Prevention Services – interventions for at-risk students (attendance or behavioral problems). Once the principal gains a deeper understanding, they should allow the district’s experts to engage the entire staff in discussions that will lead to a greater understanding of the data collected. Finally, the staff would collaborate by pinpointing appropriate intervention strategies for struggling students.