The Ultimate ESS IA FAQ

Recently, I wrote a post about how to do the ESS IA step-by-step. There are so many questions that all students will raise from the beginning to the end of the process, and here I've compiled some of the main ones. To get a feel for ideas on what to do your ESS IA on, what you should avoid or include, and if your data has been presented in the right way, you've come to the right place!


Frequently Asked Questions about the ESS IA

Should I do a questionnaire or experiment?

As an ESS student, the difficulty of your ESS IA is nearly entirely determined by the essential question: Questionnaire or Experiment?

I probably spent the most time on my ESS IA in comparison to my friends, and that's not necessarily a good thing. I was pretty much forced into a position where I had to do a questionnaire (as opposed to an experiment) due to me having only Humanities based knowledge and, because of this, no knowledge of how to even stage an experiment if I wanted to. The rest of my friends were science kids, so they all chose the experimentation option.

One major flaw within ESS already becomes apparent when doing the IA with a questionnaire: ESS IAs are largely dependent on identification and manipulation of Independent and Dependent Variables (IVs and DVs), of which simply don't exist in questionnaires since you're not manipulating anything, just asking participants questions.

On top of that, experiments have relatively less to talk about in terms of limitations and weaknesses, which you are required to discuss in the IA. The reason they have less to talk about it because these limitations and weaknesses are due to small errors on your part or general issues that were faced when conducting the experiment.

In comparison, questionnaires have ALL sorts of implications that are not only a result of your own errors, but those of the participants in the questionnaire, dozens of biases that can influence how they respond to questions, your own poor judgement in understanding how to effectively curb the influence of these biases, problems relating to the age/demographic/geographic region of the participants, etc. So, an IA just over 2000 words is simply a tiny, tiny, word count to fit that all in. Therefore, if you have a satisfactory amount of knowledge as to how to conduct an experiment, I'd probably recommend you do that over an experiment. Don't panic if you're a humanities student though, questionnaires aren't impossible but you just have to be prepared for a potentially larger workload.

How do I find a topic?

Finding an ESS topic is really the make-or-break moment of the whole assignment. It really determines how much time you spend on it and how difficult data analysis will be. 

First, take a look at if a questionnaire or experiment would be better suited to your skillset. This is essential since the amount of knowledge on either one will determine how difficult the IA is. 

The second step to finding a suitable topic is looking at your everyday life. You will be looking at certain developments in your local region (ie. the creation of certain buildings, the chemicals in waterways, the animals in your region). You will be largely deriving your experiment based on what is happening to your community right now, since that is the only region you will be able to access.

I got inspiration for my ESS IA through trying to answer a question that is relevant to my region, in that the price of vegan foods is much higher than those that aren't. I wanted to find out if these prices influenced the opinions of people in the area I live. 

Read the info below to get a better idea of how you can use your region for inspiration for your study!



What is an "environmental issue" considered in ESS?

Environmental issues are categorised into 2 major types: Local and Global.

Local issues can be specific to and inspired by what you see in your everyday life, including seeing more chemicals in your local lake, planned deforestation in the forest near your house, changes in the way the water tastes in your apartment block, etc. There are many things within our everyday lives that are somehow interacting with the environment around us, and to find a local environmental issue is to find something relevant to your own life.

One girl in my school found that when she was scuba diving, the water around the local power plant was quite warm, whereas elsewhere it was pretty cold. Based on this, she decided to take little samples of coral and seaweed from across a designated area around the power plant and then expanded her sampling to include areas further and further from the power plant. She ended up finding that the corals around the power plant grew to be a completely different size from those that were further away. She ended up actually winning an national competition with her research! This is just one of many where you can look at local developments and see what could be affected to get some inspiration.

Global issues are more widespread, but this doesn't mean they aren't just as specific as local issues. Global issues largely focus on certain local trends that are seen globally. These can be related to certain policies being implemented on a large scale due to a certain directive (such as those approved by the EU), the issues that come from increased amounts of chemicals being used in irrigation, environmental issues stemming from people not having enough money to live more sustainable lives, etc.

These global issues are usually seen as more far-reaching and easy to do, however this is not always the case. There are plenty problems that result from you only having a certain amount of resources to investigate a globally experienced phenomenon. To what extent is what you're studying giving insight into the named global issue, or does the area you live prevent you from getting actual insight into the true breadth of the issue?

Too many times do students pick a topic that is too widespread to truly investigate, whereas it would be much easier to look at your day-to-day life and understand what globally experienced issues are influencing your local environment.

What kind of responses should I have for my ESS questionnaire?

If you're doing a questionnaire for your IA, you will at some point struggle with figuring out what's the most ideal response type you should have. Deciding which of these is most appropriate is difficult, and mostly dependent on what you need from your study.

Here, I've compiled a list of response styles, what they're used for, and the potential problems you may have with each.

Numeric Scale (linear scale, multiple choice, drop down menu)

What is this? This is a scale used to understand how strongly someone leans toward a certain opinion/behaviour. Specifically, Likert scales are widely used for questionnaires and range from having 5 to 7 options, with a neutral option in the middle to avoid participants being forced to choose an opinion (this is known as Forced Choice Method). Do some research on the Likert Scale and if you find this is the right type of response style for you, definitely include the original name to sound extra fancy and knowledgeable instead of just including "a scale".

When should this be used? This is ideal for asking opinions of participants.

What problems come up when using this? This scale has major issues in terms of presenting the data. You have to present your data for either agreement or disagreement, and on different parts of the scale. Therefore, comparing results on a scale like this is pretty difficult.

Open-ended questions

What is this? This is a qualitative way to look at the issue being studied. This gives you an opportunity to see the topic through a more holistic perspective, and you can even get the reasoning behind certain actions or opinions from participants.

When should this be used? When you want the reasoning behind certain beliefs or actions.

What problems come up when using this? This is one of the most problematic methods of getting responses in an IA. The results are essentially incomparable, meaning that conducting analysis and arriving at a justified conclusion is particularly hard (especially with such a small word count). Unless your IA is entirely based on the reasoning behind one's beliefs or actions, I would opt for other response types since this is incredibly difficult to analyse, control for biases, and come to an accurate conclusion about.



How do I process my raw data?

Raw Data is incredibly complicated usually and finding the best way to process this data so seeing what all of it means is a frustrating process, particularly for questionnaire studies.

For experiments, use the exact same processing techniques that are seen in your chemistry/biology/physics classes. In addition to this, for a specifically ESS related assignment, you need the inclusion of line charts, bar charts and pie charts where possible. These charts should generally be included where you have information about the region your experiment took place, the different conditions in your experiment, or if you have different species of organism involved in your study. Anything that can be easily put into a chart should be in one!

 Questionnaires aren't the same as experiments in the way that it's quite simple to establish if something is statistically significant or not and you can move forward from there. Most often, students doing questionnaires won't even do correlational analysis, so identifying any and all trends is of utmost importance. For questionnaires, the easiest way to process raw data is by colour coding or by percentages.

I originally processed my data through colour coordination of people's responses according to how strongly they agreed/disagreed with certain statements.

Here's how I did it:

First, I needed to identify what I was researching in simple terms: the relationship between people's income and their opinions.

I needed to colour code their opinions, and as such they looked like this in my spreadsheet of data: disagree, neutral and agree.

After I had figured out a colour coding system for participants' opinions, I had a very colourful (and useful!) way of looking at my data so I could identify trends more easily. This is how all of the responses to the statements in the raw data looked.

This was very messy to me and there were way too many different colours for me to wade through to understand what happened in the responses. So, I organised the opinions further by categorising people based on their income.

I split people's income into 3 sections: low income, middle income and high income.

From here, I organised them into groups, which I could start seeing trends from much more easily:

Looking at this table, you can see how:

• Those categorised as low income seemed to disagree with many statements
• Those middle income participants were more neutral
• High income participants agreed more with the statements

From here, I could start writing down what trends I saw in a visual way. This won't necessarily be the table you'd include in your final IA but it can provide insight into what kind of things you could discuss.

When doing a questionnaire, you need to also look at participants' basic information, such as where they live, what age they are, what their background is, etc. These could all be essential in your analysis of results, since someone from the countryside might respond very differently to your questions when compared to a city dweller.


I further looked at:

• how the city participants lived in may have affected the results
• if there was a majority of participants from a certain region. 
• the age of participants and if certain important age groups aren't accounted for (this is particularly relevant if your participants are mostly 16-18 year olds and you have no older people, for instance)

You can put these details into pie charts showing the distribution of your participants, and shows to the marker that you've thought about how the background of participants may impact the study. 

What kind of figures do I need for my ESS IA?



You can include:

• pie charts
• line charts 
• bar charts

Students tend to lose points for using the wrong way to present their data, or presenting the wrong data. Students tend to just throw in many figures for different data, but they aren't actually necessary in understanding the data you've gathered.

For instance, picture yourself investigating Subject X. Subject X is about the opinions of students on meat consumption in school. It's not necessary to have 5 different pie charts about the regional differences of participants because it's not really necessary in understanding the data. However, it might be more relevant to include more pie charts about the ages of participants and their subjects. If those taking subjects about Biology and ESS eat less meat than those taking PE, then that might be more relevant for your figures for Subject X.

Keep in mind you can literally lose points if you present unnecessary figures, since it becomes confusing for the marker and makes it look like you don't actually understand what's important in your data and that you can't separate from what isn't.

---------------

That's it! You've come to the end of my post on Frequently Asked Questions about the ESS IA. This is definitely not an easy assignment you can get done in a night, so feel free to look around on my website for my other posts on the assignment to get some insight into how you can do it. Otherwise, keep checking into this website for regular posts on other assignments and don't forget to comment if you have questions or suggestions for future posts :)