Choosing a Topic
- Pick a topic that you can link with something globally significant
One of the more difficult criteria for many students is "personal engagement", in which you need to justify your experiment. I recommend picking a topic you can link with something globally signifcant - afterall, you are a student in the International Baccalaureate
For my IA, I linked my overall experiment with climate change. I argued that the two grains I used, rice and maize, are vital for human sustenance, and that by testing their respiration rate at different temperatures, I was modelling the potential effects of climate change, especially since enzyme denaturation is irreversible.
I argued in my lab report, "After the neolithic revolution, humans have become more and more dependent on agricultural yields of starch-based nutrients, such as maize (Zea mays) and rice (Oryza sativa). This has led to a selective breeding of a select species of plants for the better part of 10,000 years. However, as Harrari (2011) notes, this domestication of plants has also made humanity more and more dependent on this select variety of plants. This entails that with natural disasters or sudden changes in the environment, there is a larger risk of losing much of the nutrient supplies humans rely on today.
An imminent threat to this is in climate change, since a relatively sudden rise in temperatures may have adverse effects on the respiration rate plants such as Zea mays (Maize) and Oryza sativa (Asian rice), which are agriculturally significant, both locally in Korea, and globally"
- Make your topic locally relevant
One of the aspects teachers and moderators seem to appreciate is if you add your own, local twist to an experiment that can be conducted anywhere. It might just be a minor change to an already existing experiment, but this small twist makes it your own experiment. It could be as trivial as choosing local plants to study, or local invertibrates.
For my IA, I deliberately chose to test the respiration of rice grains, which are local to Korea, as this is locally relevant in Korea, where rice is the major staple.
- Choose a topic that renders quantitative data.
Yes, in theory, a model, or a qualitative study can earn you all the marks you need. However, in practice, this is very difficult, as teachers, examiners, and moderators are looking to tick boxes, and if you can get numbers for an experiment and draw a line graph, this gives you a lot of opportunities to look analytical without making such a big fuss.
For my experiment, I could easily test the amount of CO2, so this was much, much simpler than having to write descriptions of qualitative data. Also, an experiment allows better evaluation, which earns the bulk of the marks - so do bear this in mind.
Preparing for the Experiment
Write the Method first
- Before you start your experiment, write a method and have it checked
One very useful piece of advice our teacher gave us was to write out the introduction, hypothesis, variables, materials, and method of our lab reports before starting on our experiment. I don't know what other IB schools do, but if your school doesn't require this, I would recommend still at least writing out your method and asking your teacher to look over it before you start on your experiment. This step is important, as many, many things can go wrong with your experiment, and a teacher looking over it can be a safety net to see if you're on the right track.
Personally, I had two alternative experiments - one of which was much more elaborate than the other. The experiment I had originally preferred was to perform Sanger Sequencing of two local plants to find and compare the genetic makeup of each and thereby (hopefully) find genetic evidence for evolution from a common ancestor. However, my teacher (fortunately) advised me that this experiment would require several expensive enzymes, so I was prevented from going towards a dead end. Writing out a method also helps you to think about any potential problems you might have and saves you time later on.
Find the Materials
- Find equipment at school and purchase additional materials well ahead of your experiment.
This is perhaps a very practical piece of advice. However, it's well worth giving, as several of my friends were unable to start on their experiments in the first class, as all required materials were unavailable. I recommend finding your required equipment, purchasing required materials well ahead of time, so you don't waste time.
For my experiment, I needed rice and maize seeds, but for a fair experiment, I realised that I needed unprocessed rice typically used for planting. After some searching, I found that the Korean government regulates the distribution of such unprocessed rice and had to contact the Korean Agricultural Development Centre, which kindly donated rice for my experiment. As for the maize, I had to purchase this from a seed shop. I had time to prepare this during a spring break, so planning ahead of time was really important.
Also, I originally planned on making an old-fashioned respirometer, but our teacher directed me to the fact that our school has carbon sensors, which give much, much more accurate readings, and are much easier to use - so I saved so much time by just asking about such equipment.
- Spend time outside class collecting data
The IB claims that the whole exploration should take 10 hours. Nobody believes that - not even the most fervent IBelievers. Show your personal engagement by spending time outside class experimenting. This is vital if you actually want to collect the vast amount of data required for a high score.
For my experiment, I spent at least 10 hours just collecting my data. Remember, at least 5 trials are needed for you to use Standard Deviation in any meaningful way, and I used 5 different temperatures for 2 different types of seed, so I collected a total of 50 pieces of data, which was extremely tedious. I spent a vast bulk of my study halls and after school time to collect all these pieces of data, which, I believe made my work stand out.
I remember that there was only one of the person that stayed behind for such a long time after school for his experiment in our class, and I know the two of us scored very well - I think in the end, the more effort you put in at this stage, the better your data will be to analyse.
- The Macro: Begin with the global and local significance
Personal interest isn't a sissy "I've always liked..." or "From when I was young, I always loved..." - it should really ring true for the real world. In this sense, the best way to imply your keen interest in your experiment is by justifying how it relates to the global and your local contexts.
A good question to ask yourself is, "Although high school equipiment does limit you, if you weren't limited, in which direction would your overall experiment have gone?"
In my experiment, I used two independent variables - type of seed and temperature to test the different respiration rates. Though I was limited by time and so could only test two types of grains, I argued in my introduction, that humans had become increasingly reliant on a handful of seeds, which, if changed, can have drastic effects on the staple products of humanity. I then justified why I chose the two specific seeds using locality - rice is a staple of Korean diet. Though I was limited to only 5 different temperatures, I suggested in my introduction, that an inherent potential problem for our species is climate change, which could lead to denaturation of enzymes, leading to irreversible damage. In reality, climate change by any means would only be by 1-3 degrees (rather than the 10 degree intervals I was limited to due to the water bath not being precise enough). However, I was quite creative and argued that a temperature increase in the extreme can be damaging (that is, if 50 degrees is the optimum, if 51 degrees is hit once on an extreme day, this could lead to denaturation, which is irreversible).
I reckon you need to think of such implications and realise that you can stretch your experiment's implications with a little creative thinking - bearing in mind that you are limited by high school lab equipment.
- The Micro: State the mechanisms that make your experiment work
Often you will have to write a whole section about enzymes (which is, by far, the most popular mechanism used in experiments). Even in my experiment, I linked respiration to enzymes - how respiration is controlled by enzymes and enzymes increase with increased temperature until denaturation. You really just need to paraphrase from your textbook at this point, though if you want to add anything extra, some scientific literature (easily available on Google Scholar, JSTOR or the like) can be the icing on the cake.
- Describe similar studies and how your experiment is a unique contribution
A good lab report should describe some similar studies, so that your experiment starts off where another ends. You should try to describe the theoretical ideas explored in similar studies and try to describe how these ideas apply in your own experiment.
A great lab report will go further than this to explain how the experiment is truly a design lab - how it contributes to scientific research through its own unique insights. You should start doing this in your introduction - justify why your experiment is worth pursuing and how it is a unique contribution to your science.
- Avoid over-speculation
An hypothesis is not a place for you to speculate too much about your experiment - it isn't about using intuition - it should be a reasoned piece of work with citations from scientific literature. It isn't cheating to look up similar experiments before your own experiment to find what other scientists have found. In this sense, you should make a reasoned hypothesis with theoretical reasoning and cite other similar studies to give reasons for why your experiment might go one way or another. You're not in an elementary school lab class, where your teacher might ask you to "just make a guess" for your hypothesis - you need to give a reasoned hypothesis at a higher level than this.
- First state an obvious hypothesis
"Especially if you followed my earlier advice and have two independent variables, you shouldn't neglect giving an hypothesis for the more obvious independent variable.
This is IB, where you state the obvious
~ An anonymous English Literature teacher
In my report, for instance, I first gave the more obvious hypothesis - that as temperature increases, respiration would increase to a certain point (the optimum) and decrease rapidly after this point, given that the enzymes responsible for respiration would denature. I used the textbook and several other sources to support this hypothesis.
- Next, dig deeper and make a less-obvious hypothesis, supported by scientific literature
Now the harder hypothesis to form is for this one, and it is often very difficult to find a good response from reading the textbook alone, so you may have to do some extra reading of scientific journals and studies.
In my case, for this part of the hypothesis, I had a much more difficult time determining which of the two types of seeds would have a higher optimum temperature for respiration rate. However, determining this was the main part of my experiment - as the implications most relevant concerned climate change and suitable grain for humanity. I read several scientific experiments to support my hypothesis and also read fairly widely about rice and maize, their original locations, and their ancestral species to support my hypothesis. For a strong, well-reasoned hypothesis, you need to read well beyond your textbook to suggest an outcome.
- Two Independent Variables is Ideal
Listen - you're not in a middle school science class - you can have more than one independent variable - so long as you don't get confused - a good idea is to have one numerical, continuous independent variable (that allows you to draw up a good line graph) and a categorical, discrete variable (that allows you to have multiple lines on said line graph). This gives a level of complexity to your experiment, which is vital if you are aiming for a really high score (a 6 or a 7). I guess you could have more, but this might be unnecessary.
- Numerical Dependent Variable
Though you can, in theory, do an IA without numbers (observations or models), my teacher very helpfully suggested conducting an experiment that renders numerical data for the dependent variable. I find this very helpful, as it allows you to more easily analyse your data through presenting a line graph. Perhaps this relates to TOK and languages in the sciences, but numbers tend to give the impression of precise data as well, as opposed to a good description.
- Systematic and Random Error
Before you write about controlled variables, you need to clearly distinguish between potential sources of systematic and random error. Systematic error is a flaw in the design of the experiment or the equipment used, such that there is a continuously higher or lower measurement of the dependent variable. As such, this type of error can (theoretically) be calculated and resolved after the experiment is conducted, but is much more serious, as it cannot be corrected for by numerous trials. By contrast, random error, as you would expect, is caused by unpredictable changes in the circumstances of conducting the experiment. As such, this type of error can be corrected for by conducting numerous trials (and hopefully the random errors that lead to a higher measurement and random errors that lead to a lower measurement would cancel each other out with many trials). You can really only control systematic errors, and the best way of controlling random errors is to conduct many trials (for the sake of IB IAs, 5 trials or more should suffice - the goal is to make the error bars as small as possible)
In my experiment, I suggested that type of Carbon Sensor used could be one of the sources of systematic error, as different carbon sensors can give higher or lower readings and therefore used only one carbon sensor for the entire experiment (thereby controlling this potentially confounding variable).
- Describe the variable and explain how the variable is operationalised in your experiment
I reckon this is the real mark-winner from this section. I was quite used to making a table for this part from when I was in MYP, but for those of you new to IB, I suggest making a table with 3 columns - the first column states the variable - the second column describes the variable - the third column describes how the variable is operationalised in the context of the experiment.
Let's take an example from my own experiment. In my experiment, I had to control the amount of time I spent for each trial. Why? Well, imagine if I gave 1 second for the maize at 25 degrees and 10 hours for the rice at 25 degrees. Then of course there would be more respiration by the rice - I gave it 36,000 times more time!
So in my first column, I simply stated, "Amount of time for each trial". For my second column, I described why this could be a problem, "The more time the seeds are given to respire, the more carbon dioxide would be produced, meaning that differences in the amount of time for each trial could influence the amount of carbon emitted." For my third column, I described how I could control this in my experiment, "The time for each trial is controlled at 10 minutes."
- Be precise in your method
As we learn in TOK, a key part of the scientific method is replication. Scientists replicate experiments all the time in order to support or disprove the findings of previous experiments. As such, your method needs to be precise enough for one of your peers to be able to replicate your experiment.
One good exercise (if you have the time) is to work with a friend and read each other's methods. Perhaps you can imagine yourself doing the other person's experiment - anyhow, the key here is to ask questions about details that might be important in the experiment. Hopefully you will realise that you are missing some details that are important to replicating your experiment. Also, take a look at your list of variables to check if you kept all the variables controlled.
Though this process is tedious, you need to ensure that your method is clear enough with enough details to follow. Bear in mind that you are assessed also for "Communication", which entails that you need to be very clear in your method.
- Add a diagram of your experiment set-up
Though not formally required, adding an annotated diagram of how your experiment was set up (either as a drawing or via a photograph) can enhance your communication. I added my diagram as an appendix.
Data Presentation and Analysis
- Raw Data to Appendix
Since the raw data does take a lot of space, just toss it in as an appendix and refer to it in your text. (i.e. For the raw data, refer to Appendix 2)
- Sample Calculation
You need to show basically how you arrived at the processed data, so show even the more simple steps (such as subtraction and getting an average) to show how you arrived at these numbers.
- Qualitative Data
Though not a formal requirement, if you can describe some data qualitatively, this is a big bonus, as it shows that you are using multiple ways to communicate your data. It ensures that you are opening up those extra marks for data collection as well, since you are also describing your data in a different way.
In my report, I wrote, "While it was not possible to directly observe the intake of oxygen and production of carbon dioxide, it was possible to notice that at 65°C, several light brown parts were visible in O. sativa seeds"
Though perhaps less relevant to the substance of the experiment, I feel that adding this qualitative data shows the teacher that you aren't skipping over this step.
- Graph with description and analysis
I recommend making your graph on Excel, as it allows for many different functions that may be necessary. Often, you need to use the "Select" button to change which set of data are in the X and Y axes. Remember to add a title to the graph and to each of your axes. Remember to add error bars using standard deviation (as necessary).
For your description and analysis of the graph, begin with the obvious - describe the graph. Perhaps helpful here is to divide the graph into two or three parts - perhaps there is a part where there is a rapid increase then a less rapid increase or an increase then a rapid decrease. Describe these trends. The less obvious part is to analyse - one easy way is to explain why this is happening in the graph (especially using knowledge from the Biology textbook and outside reading).
- Reliability vs. Validity
As with random and systematic sources of error, reliability and validity are really important to distinguish, so you can write accurately about both in your analysis of the data.
Reliability is about how precise your data is. In other words, reliability has something to do with the size of your error bars - whether any overlap - whether they are large or small. A low level of reliability is most likely due to random errors and can be remedied through multiple trials.
By contrast, validity concerns whether the data is what you set out to attain. It has something more to do with your method and whether your method renders what was intended in your aim and hypothesis. It also has something to do with sources of systematic error - if there was a systematic error that ran across your experiment, this would hinder the validity of your data.
For a strong report, you will need to write about both. The challenge is to be critical of your own work - but perhaps think of any unreliability not as a fault of your own, but caused largely by the contstraints of time and highschool equipment. Be as critical as you like about the work you did, but balance is required here - if you're too critical, this puts your method into question - if you're too lenient, this is not good either. Balance is required.
- Practise for Paper 2
Data analysis is perhaps the most important part of your report. To be brutally honest, this was the part I personally procrastinated on the most. In hindsight, I think you should treat this part very much like a Paper 2 - which is pretty much data analysis anyway - literally time yourself and write a response on paper as if you would write a Paper 2 and make this the start of your analysis. Of course, a 30 minute handwritten analysis needs to be polished, but this might be a good way to kill two birds with one stone - prepare for Paper 2 and write your IA at the same time!
- Think about both elements of your hypothesis
In fact, you need to restate your hypothesis in your conclusion and state whether it was supported or not supported by your experiment. Explain why this conclusion was reached. An excellent report will also include to what extent the conclusion can be reached - explaining that the reliability of the data may make the conclusion difficult to substantiate, for instance.
- Support with Scientific Literature
This part shouldn't be too hard if you have read widely enough for your hypothesis. Compare and contrast your work with that of other scientists and where possible, support your own findings with scientific literature, including citations.
- State the implications of your experiment
Though not formally required, I simply restated the implications as laid out in the introduction and suggested how my experiment fits into the real world.
I wrote, "The implications for these results with the effect of climate change could not be discussed effectively because although there has been a minor increase in average global temperatures as noted by NASA (2014), these average temperatures do not provide for the local extremes in temperatures. Simply taking the average temperature into account, a 0.8oC increase would have a positive effect on the respiration rate of these two plants, since the average global temperatures would not pass 55-65oC, which was identified as the optimum temperature for respiration in rice. However, the overwhelming majority of recording devices for climate, are kept out of direct sunlight, whereas crops such as rice and maize are in direct sunlight, as criticised by Nicolls et al. (1996). This could potentially mean that an extreme temperature in a particular region, could irreversibly damage the supply of crops. Because the process of denaturation cannot be reversed, the threat of dangerous levels of human-induced climate change could have adverse effects that cannot be foreseen by looking only at average temperatures."
Evaluation and Extensions
- About three of each should do
Again, a balance needs to be reached, as you don't want to give the impression of a method with too many holes. Think about the three biggest potential sources of error and about three most interesting, potentially insightful extensions given your implications. You should try to describe each of the evaluation and extension ideas in a great amount of detail (about 3-4 sentences).
- Address error bars as necessary
For most people, the ideal conclusion is not able to be reached because of large error bars. Here is a good opportunity to address the sources of errors that may have caused these. Several sources of errors can be attributed random factors such as the inability to perfectly control confounding variables, the inability to perfectly measure the dependent variable, or the inability to perfectly control a continuous independent variable.
For instance, in my report, I gave two random sources of error - the fact that temperature could not be controlled perfectly in the water bath, and was often one or two degrees above or below the stated temperature - and the fact that the carbon sensor had different margins of error for its two settings.
Also, you should address any systematic errors, as appropriate.
In my report, I explained, "The seeds could not be measured at exactly 100 grams. Since it was beside the point to cut seeds into smaller pieces merely to fit the controlled mass, it was decided to go a small amount over the 100 grams, rather than slightly under. However, because in rice, the outer layer does not require respiration, but counts in the mass, the mass should be replaced by the number of cells that conduct respiration"
Notice how I also provide a solution - how the potential errors could be fixed in a perfect world with better equipment and time.
- Think back to the implications of your experiment to explain how to extend your experiment
Here, try to imagine that you are a professional biologist and have access to more resources and more time. Suggest ways you could extend your experiment, bearing in mind that your new ideas should be linked in some way to your experiment. Think about the reasoning you gave for why the experiment went the way it did and if there are ways to test these.
For instance, in my experiment, I wrote, "It was speculated, albeit with reason, that there are different enzymes in the respiration of Z. mays, compared to O. sativa. It would be worth finding exactly what chemicals differ, and this could be done through grinding and using a centrifuge to separate the elements of the two types of seeds, and then conducting a series of chemical experiments to find what chemicals are present in the two types of seeds".
Notice how I am explaining that there is a definite link between this new, extension and my own experiment, which could aid me in forming a stronger conclusion. This is what an extension should do.
After your Lab Report
Preparation for Paper 2
- Make data analysis questions for your lab report and switch it in a study group
Something that I found very useful when I was tutoring several students in Biology after completing IB was to show them the graph made from my Lab Report and asking them some critical questions about these. After giving some context, I would ask questions like, "Describe this graph", or "Calculate the difference between...", or "Suggest reasons for these results", or "Evaluate this experiment", all of which are very much Paper 2 style questions.
I find that rather than the artificial questions in the textbook, which students may feel some distance from, asking questions on each other's experiments is something scientists do all the time - so it is something well worth doing in preparation for a paper - and in preparation to become scientists.
This activity is something I might recommend to students in study groups and to IB Biology teachers, as it is something that would have been really interesting, especially from my perspective as a Bio-loving student.