Subject report and IA
Thursday 1 March 2018
Preparing for May 2018 Submission
This post is about some points to consider before people submit their May 2018 Internal assessment marks and samples. It is always a good idea to pay careful attention to the subject report from the last year. This is not always written in the friendliest of terms and doesn't always seem to be sympathetic to the challenges of teaching this group of students. It often seems like an expression of the frustration that examiners have and un realised potential. Many of us, I am sure, would appreciate the chance to express some of our frustration at the moderation process. That said, we must remember that we all have a common goal and we do get our chance to feedback on the assessment process and teacher returns are startlingly low. Also, the net result of the subject report is very clear advice on how to help our students - the common goal. for that reason, we should pay careful attention and try ti implement these ideas. Before I go further, let me be clear that I recommend teachers visit their MYIB account and read the appropriate report/s in full. What follows here are just a few important highlights about the IA only.
My Take on Additional points
It is worth being clear that the examiners report should be used as a source of additional points to consider when assessing and that adherence to the full marking criteria and guidance in the TSM is the main reference point for assessment. I will pay particular attention to the following points since they have been specifically raised last year, but this is in no way a replacement for the full marking process.
ALSO - the following is just my take on the points that are raised and is not offcial IB feedback. read the report yourselves. I have just tried to offer helpful related advice.
Please note that there is an extensive section on the Internal Assessment on this website with huge amounts of details, ideas and examples to help with this process.
Key points about IA
Data/information
I thought it was quite nice to read a specific mention of the idea that a third variable offers good depth and potential to a project. The report specifically states that projects are 'Strengthened by a third variable'. This is fuel to the idea that I talk about a lot on workshops that we might encourage students to work on 'What factors affect suicide rate?' as opposed to 'Is there a relationship between suicide rate and GDP' (for example - I am sure you can think of others). As such students have more of a 'theme' to investigate instead of a 'question' to answer. This does not necessarily imply a big increas in size, but just that there might be an extra process or two to add to the discussion and provide a bit of insight. This does not mean that there MUST be a third variable, but that awarding 3/3 in this criteria might depend on one.
The report also mentions that often the information collected was too sparse to acheive the stated goals or to use some of the processes.
POINT - Consider the award of 3/3 carefully. If the candidate simply has two variables with no other way of splitting the data up then this might preclude a 3.
FOR NEXT TIME - I give students some guidelines - For each 'thing' you are collecting data about, try to find 3 numerical data fields and 2 categorical ones. This ensures potential for fruitful investigation. Check the Internal Assessment section for more details.
Processes
With [ marks at stake, this is often the hardest. Many of the usual subjects showed up here.
- Relevance - This is repeatedly mentioned. Less is better. Short and concise projects with simple processes that are appropriately used to answer the question are key. Too often, repeated tenuous number crunching can limit a candidate to 2 here.
- Defining variables - this has come up before and it is important to pay attention to this. Students often overlook this because they are so involved in their projects. Teachers can do this too. remember that it will be read by a stranger who does not know you, the student or the project. Examples might include - being clear about what x and y represent on a scatter graph (define the dependent variable when possible too). When substituting in to a PMCC formula, all of those variable need to be defined. If a student writes \(\overline { x } \), they should be clear about what this means and about which variable they are talking. the reader shouldn't ever be asking 'what is this?'
- Calculations by hand - this is intriguing, since we are told that processes using technology only are considered simple, but are now also being told that students should be encouraged to show hand calculations even when technology has been used. This is a potential contradiction but I think it is about the frequency with which calculations just appear in projects with little or no explanation. Students should show awareness of the meaning of the calculation I think to counter this. That said, I will put a bit of extra emphasis on this for this year's sample to show willing. Where posisble I will ask students to say something about the 'mean' and 'SD' calculations. In general, I will encourage students to explain about the process used to demonstrate their understanding of it.
- Explaining screenshots - This is an interesting one. The implication here is that candidates are pastuing screenshots of spreadsheets, GDCs and stats apps in to their projects expecting the reader to know what is going on. I think it is a valid point. Again, it stems from people who are so involved in the project they don't stop to think about it. Annotations, defining variables and so on to be encouraged here.
- Use it! - For years we have been battling the random insertion of a regression line on to a scatter grapgh (often where it is not justified). Now, even where it is appropriate, the case is often that students don't use it for something. As a rule, if a process and its result is not discussed or used then it is arguably irrelevant. I imagine that this is pretty frequent and easily missed by busy teachers in the marking process.
POINTS - Mathematical processes should be simple, clear, well communicated and relevant to the stated aims and the result should be discussed and used. These seem like sensible aims!
FOR NEXT TIME - I do still put a lot of effort in to the planning phase. My aims is that students do actually have a very clear and coherent plan before they actually do any mathematics. Sure, perhaps that will change after they have, but I think it is very important to think coherently about this first. I also put a lot of stock in asking students to ask an outsider to read their project critically. This often throws up some of these isuues
Interpretation
This has always been tricky. Some of the example projects that are given 3/3 here are not very complex or sophisticated and yet the criteria are clear about meaningful discussion. Ogf particular interest to me this year is the comment that 'Candidates should be discouraged from making unsubstantiated conjectures'. Clearly I get that, but I think drawing potential inferences from results is a key part of reaching the next level. I have an issue with the unsubstantiated conclusion, which I see all the time, but the conjecture I think is more relevant. It is important to speculate about possible causes for results so as to inform possible future investigations. never the less, I will pay careful attention to this and make sure that where students have done this, they have been clear about its speculative nature.
Validity
I have seen a lot of projects as both a teacher and a moderator and do sympathize with the points made here. Students will often simply state that they think their work is valid, when the real aim of this criteria is that students look for the weak links in their work. This is just science and is a process that should be encouraged across the DP. It is very much the ToK element of the exercise and should appear in that context; The project will have limits and flaws as almost all research does and students should get good at looking for it. Commenting on the data is easiest and most common, but students should be clear about how that limts the validity of their conclusions. For example - is a particular part of the population under represented? Less easy/commin, is discussing the validity of a process. For example, a student may have used a regression line where a curved model might have been more appropriate (100m sprint times are a good one here. The data appears to fit a linear model, but clearly, over time, this implies a possible time of 0s). It is a case of awareness and observation, that is all.
POINTS - Students should start from a position of assuming their projects are flawed in some way and have limits. the exercise is to find and explain them.
FOR NEXT TIME - Use some specific examples as a ToK moment in your class. Use this years projects as real examples and get students to look for the flaws and limits. Tell the ToK teacher you are going to do it - better yet, ask if there is a slot in the ToK scheme for this kind of exercise. Good for everyone.
Structure and communication
Comments here are often a long a similar theme. Again, I advocate the giving of projects to an outsider for insight on this. To get 3/3 here it has to be really coherent and not leave the reader asking questions. The Student writing guide offers lots of specific advice about this. It is also hard to get full marks here if the project is too simple. I know that is subjective, but there has to be a stroy to tell if you want to tell it well. The aim is to thread the whole thing together really well.
I was particularly intrigued by the mention of Chi2 categories here. This is problematic for lots of reasons, but I was interested to see it mentioned under this criteria. I have written before about the idea that chi squared tests are designed for categorical data and that if both variables are numerical then candidates should sue a scatter graph. Clearly though, if just one of the variables is numerical, then the chi squared test is still needed and the numerical data has to be split into categories. How this gets done is a real issue for me. My conjecture is that the categories can be defined as you please to fix the outcome. (There is a bold conjecture and plenty to discuss in the validity section). As such, if candidates are unable to justify their choice then I think this has to be addressed somewhere in the criteria. I'd be happy to do it here as suggested by the report rather than in criterion C. Suggested ideas for categorizing are...
- Use quartiles - Below the lower, the interquartile range, above the higher - but acknowledge that this will fix your frequencies (that is OK, but just needs acknowledging)
- Use means - Above and below the mean can work well, but beware that it offers only 2 categories and so unless you have more than 2 for the other variable you will end up with a 2 x 2 and need Yates's continuity correction.
- USe existing definitions - This depends on their existence, but for some variables it is likely that candidates will be able to find some boundaries that have been drwan up 'Below this is considered low' for example
POINTS - Candidates should make sure they explain the decisions they have made. The reader should not left lost or uncertain at any point on the project if you want to award 3/3
FOR NEXT TIME - I think I will do a specific activity on splitting up categories for chi squared tests so that students have experience to draw on if and when this comes up.
Notation and Terminology
Again, the usual comments about computer notation have come up here and they do need careful attention. The main point I take from this report is the advice that students are explicitly shown how to use an Equation editor. It occurs to me that I have not actually done this. Although my students do pretty well here, I think this is a good ide
FOR NEXT TIME - Think about how much help we give students wit this level of mathematical communication
More feedback
There were also some general comments about helping students to understand the criteria and suggestions that we could offer students more guidance. This is good to hear as I know teachers often worry about this. What the report often misses though is any appreciation for the number of students we teach who make themselves unhelpable. You can give all the guidance you want to them, if they don't take any notice of it, then we will end up submitting a poor project. I sometimes bristle at the implication that this is because of a lack of guidance. My Student writing guide it is pretty clear and every year I have a handful of students who produce fabulous work and I can tell they have read and acted on advice. The same is true of some of the weker work, but every year I have students who repeatedly fail to act on advice given. What can you do?
Annotations
Finally, there is specific mention of examiners preferences for annotations and justifications. It is explicitly mentioned in a number of places that they want to see annotations ON the student work that show EXPLICITLY the EVIDENCE that you are using to justify the marks given. Also there is a reminder that, at the examiner end, the result is a static on screen image so teachers need to check the final uploads to be sure everything is clear and visible.
I have written in more detail about Electronic Uploads including an example.
Good luck with your marking this year. I try not to worry too much about it, but I know many of us have had odd blips in the moderation process and we all want the best we can giuve for our students. If we do our best to guide them and to pay attention to the details when we mark and submit the projects, we have to sit back and let the system do the rest.
Hope this helps!