Travis DixonAssessment (IB), Internal Assessment (IB), Research Methodology

Hypotheses can be tricky. I hope this blog can help make them a bit easier.

Updated June 2020

Writing good hypotheses in IB Psychology IAs is something many students find challenging. After moderating another 175+ IA’s this year I could see some common errors students were making. This post hopes to give a clear explanation with examples to help with this tricky task. 

Null and Alternative Hypotheses

Null Hypothesis (H0)

Our teacher support pack has everything students and teachers need to get top marks in the IA. Download a Free preview from https://store.themantic-education.com/

The term “null” means having no value, significance or effect. It also refers to something associated with zero. A null hypothesis in a student’s IA, therefore, should state that there is (or will be) no effect of the IV on the DV. This is what we assume to be true until we have the evidence to suggest otherwise.

A common misconception is that the hypothesis is based on the sample in the study. Our hypotheses should actually be about the population from which we’ve drawn the sample, not the sample itself. Therefore, when writing our hypotheses we can use present tense instead of future tense (e.g. There is instead of There will be…).

Having said that, in the IB Psych’ IA, the IB is apparently assuming the hypotheses are based on the sample (because variables need to be operationalized) so writing your hypotheses as predictions of what might happen in the experiment is fine (see below for examples).

IB Psych IA Tip: It’s fine (and even recommended) to state in your null hypotheses that there will be no significant difference between the two conditions in your experiment or any differences are due to chance (see footnote 1)

The Alternative Hypothesis (H1)

This is also referred to as the research hypothesis or the experimental hypothesis. It’s an alternative hypothesis to the null because if the null is not true, there must be an alternative explanation.

Generally speaking it’s not a prediction of what will happen in the study, but it’s an assumption about what is true for the population being studied. But, similar to the null hypothesis in the IB Psych IA you can (and should) write this about a prediction of what you think will happen in your study (see examples below).

This must be operationalized: it must be evident how the variables will be quantified, and may be either one- or two-tailed (directional or non-directional).

Read more: 


To avoid issues with copying and plagiarism, the following examples are from studies that students cannot do for the internal assessment. Some are taken from this post on how to operationalize definitions of variables.

A Fictional Drug Trial

  • H1: Taking Paroxetine  will decrease symptoms of PTSD.
  • Ho: Taking paroxetine will not decrease symptoms of PTSD.

Operationalized (as if for an IB Psych IA):

  • H1: The experimental group who take 20mg of Paroxetine (as a pill) every morning for 7 days will have a larger decrease in symptoms (as measured by the CAPs scale) when compared to the control group who will take an identical placebo pill every morning for 7 days.

A Fictional Study on Body Image*

  • H1: Viewing media that portrays the thin ideal increases feelings of body image dissatisfaction.
  • Ho: Types of media viewed does not affect body image dissatisfaction.

Operationalized (as if for an IB Psych IA):

  • H1:Watching a video portraying the thin ideal in a Baywatch film trailer will result in higher scores on the Body Shape Questionnaire (BSQ-34) compared with watching media with “normal” body types in the Grownups film trailer.
*This entire IA exemplar is included in the IA Teacher Support Pack. 

A Fictional Study on Weight Training

  • H1: Listening to music affects training performance.
  • Ho: Music has no effect on training performance.

Operationalized (as if for an IB Psych IA):

  • H1: Listening to heavy metal rock music (AC/DC songs) causes a difference in the number of push-ups performed compared to listening to classical music (Mozart’s symphony #41).

One vs. Two Tailed

It is important to know if your hypothesis is one or two-tailed. This will influence the type of inferential statistics test you use later. If you have a one-tailed hypotheses, you should use a one-tailed test. And if you have a two-tailed hypothesis? You guessed it – a two-tailed test.

The one vs two tailed debate still continues in Psychology (read more). The IB ignores this and makes it simple: one tailed hypotheses = one tailed test. No ifs, ands, or buts!

If you are predicting that one of your conditions in your experiment will have a higher value than the other, it’s one-tailed (because you know the direction of the effect – the IV is increasing the DV). Similarly, your hypothesis is one-tailed if you are predicting that manipulating the IV will cause a decrease in the DV.

However, if you think your IV will have an effect, but you’re not sure if it will increase or decrease it, this is two-tailed.

Of the three examples above, can you tell which one is two-tailed and which one is one-tailed?

Operational Definitions

Read more about operationally defining your variables in your hypotheses in this blog post.

Points to Remember

  • Hypotheses are based on the population, not the sample, so you can write in present tense. However, the norm for IB Psych IA’s is to write in the future tense as a prediction of what will happen in your experiment.
  • In IB IA’s, we’re hypothesizing about a causal relationship of an IV on a DV in a population – the hypotheses should reflect that causal relationship.
  • Inferential tests are test of the null hypothesis (hence it’s called null hypothesis testing). We are conducting the tests to see the chances of obtaining our results even if the null is true (i.e. there is no effect).

Footnote 1: Saying “that there will be no significant difference between the two conditions in or any differences are due to chance” is technically an incorrect way to state a null hypothesis. That’s because when we conduct our inferential tests we’re seeing what the probability is of getting our results even if our null were true. So if we get a p value of say 0.10 (10%), according to the above null hypothesis we’re saying there is a 10% chance that there will be no significant difference between the two conditions, which isn’t actually accurate (don’t worry if I’ve lost you – it’s mind bending stuff). This is one of those instances where poor statistical practice has ingrained itself in IB assessment. But on the plus side it does make it easier for students (and not enough time is spent on this for the bad habits to be too ingrained anyway).