Blog entry 5: Hypothesis testing

In this blog, we go on the expand further into the next part of DOE. After collecting data, we need to perform hypothesis testing, in order to determine if our initial prediction on the effects of different factors on the flying distance of the projectile is correct.

I will be using Run #1 and Run #3 to determine the effects of projectile weight on the distance travelled.

From our last practical, the data we collected is:

Scope of the test:

The human factor is assumed to be negligible. Therefore different user will not have any effect on the flying distance of projectile.

Flying distance for catapult is collected using the factors below:

Arm length = 28 cm

Projectile weight = 0.86g (Run #1) and 2.0g (Run #3)

Stop angle = 55°

State the hypothesis:

Null hypothesis (Ho): Heavier and lighter projectiles travelled the same distance.

µ1 = µ3

Alternative hypothesis (H1): Lighter projectiles travelled further than heavier projectiles.

µ1 > µ3

Analysis plan:

Sample size is small, only 16 data points (8 data points each run, 2 runs), hence the t-test is suitable.

Sign of H1 is > hence a right-tailed test is used.

Significance level (SL) = 0.05

Calculate the test statistic:

	Mean (cm)	Standard Deviation (cm)
Run #1	213.3	6.17
Run #3	195.2	3.33

Conclusion that answer the initial question:

Using a lighter projectile results in further distance travelled by the projectile, while using a heavier projectile results in shorter distance travelled.

Compare your conclusion with the conclusion from the other team members:

Jeremy Tan (as Thor): Using a lighter projectile results in further distance travelled while using a heavier projectile results in shorter distance travelled. (Similar to my conclusion)

Yan Zhen (as Captain America): Using a smaller stop angle results in further distance travelled while using a bigger stop angle results in shorter distance travelled. (Similar to Jeevan's conclusion)

Jeevan (as Black Widow): Using a smaller stop angle results in further distance travelled while using a bigger stop angle results in shorter distance travelled. (Similar to Jeevan's conclusion)

What inferences can you make from these comparisons?

- Using a lighter projectile will results in further distance travelled compared to a heavier projectile.

- Using a smaller stop angle will also results in a further distance travelled compared to using larger stop angle.

- Hence in order to achieve maximum flying distance, we should use a smaller projectile with a smaller stop angle.

Learning reflection:

There is no practical sessions this time, as we used the data collected from our last practical to learn how to perform hypothesis testing on them. Despite the lesson being a bit boring and technical, it is still important, because as we make the prototype, we need to be able to determine if our predictions on the effects of certain factors are correct.

As Mr Chua explained the concepts and equations used to perform hypothesis testing, it does get very confusing very fast for most of us, as we were introduced to completely new equations we have never seen before, as well as the Distribution Table. The best way to learn how to use these new information is to practice, and that's what we did. We were given various exercises to practice different variations of scenarios we might encounter when we collect real world data.

Overall, for me, hypothesis testing is quite a challenging thing to grasp, after the lesson, I only have a vague understanding of how it works, not much more than remembering the steps. I think I will need to practice and read up on this further. But despite being challenging, I o find that this is a useful tool which I can see myself using for future projects.