The scientific method is one of many ways to learn about nature, the universe, and ourselves. It was first formalized around 500 BC by Aristotle and then developed further over the next few centuries. Since its inception, it has been adapted and modified for different purposes depending on what questions scientists want answered.
The process goes like this: researchers must have an idea they wish to test. This could be due to a question they are curious about or because someone told them something that seems plausible. Then, they must gather some materials and tools needed to conduct their research.
Next, they must determine how well each material or tool correlates with their hypothesis. They may do so through direct experimentation or indirect comparisons with other information.
Once all the components are gathered, the researcher can perform his or her experiment. During this stage, the researcher will be gathering data either qualitatively (by asking questions) or quantitatively (by measuring things). After collecting enough data, the scientist compiles the results into a thesis or conclusion based on his or her hypotheses.
After confirming his or her theory, the researcher elaborates upon his or her findings and compares them to similar theories.
Test limitations
A key limitation of most statistical tests is that they only work if there are enough data points to determine whether the hypothesis being tested is true or false.
That doesn’t seem very intuitive, does it? After all, wouldn’t it be great if we had a test that could tell us with 100% certainty whether something was true or not?
Well, we do have such a test, but it relies on looking at how much matter there is in the universe. That sounds pretty speculative, I know!
Fortunately, science has a way to deal with things that are too uncertain. They call this process inference. By inferring what you think about the world based on everything you already know, you can come up with an estimate for the likelihood of a particular event occurring.
This process — often referred to as probability theory — is used in lots of areas, from medicine to economics to psychology. And while it may sound like more mathematics, it really isn’t.
Instead, it uses logic to evaluate assumptions made about the world. Probability theory simply teaches you how to use logical reasoning efficiently.
In fact, many people consider using probabilities as a second language. So even though math is involved, that isn’t too difficult. But now, let’s see how we apply this concept to scientific research!
The example given here applies directly to our topic today –- testing new treatments for depression.
Sample size
A key part of any scientific study is determining how many people will be involved in it, and what conditions they will be under while being studied.
A small sample size means that there are not enough participants for the experiment to get solid results. This could mean no clear conclusion or even false conclusions because there was not enough data to make assumptions about the rest of the population.
If you want to do scientific research, you must determine your minimum needed participant number carefully!
By having a large enough sample size, researchers can draw more definitive conclusions from their studies. Make sure to weigh the benefits of having a smaller sample size against the potential lack of accuracy when doing experiments!
It is important to have an adequate amount of participants for your experiment to work properly.
Statistical significance
What is statistical significance? That’s probably the most difficult part to remember! But it’s very important to understand, so let’s take a closer look.
Statistical significance refers to whether or not there was actually a difference between two sets of results. In other words, if you can’t determine with confidence that one set of results were better than another then there isn’t really a reason to change which set of results you use as your baseline.
This is why researchers use both their own experiences and those of others for comparisons. For instance, they might compare their experiment’s outcomes to what has been shown to work in past studies, or they may choose to use something like the average value when talking about how theirs differed from the norm.
It’s also worth mentioning that even if a study does find an effect, this doesn’t mean that its intervention worked exactly as planned. A lot of times, things don’t go according to plan because no matter what people do, some individuals are just genetically predisposed to be susceptible to certain diseases and treatments.
Confidence intervals
A confidence interval is an important way to determine how confident researchers are of their findings. It comes from probability theory, where you have a chance or likelihood of something happening. In scientific research, there is a statistical term for this chance — called “confidence.”
So, when scientists perform a study, they use a certain level of confidence to tell them whether their hypothesis was true or false. They may hypothesize that drinking eight glasses of water per day will help you lose weight, so they test this theory by giving ten people either eight glasses of water daily or a bottle of water every night before bedtime for one month.
If the results show that there is no difference in weight loss between the two groups then we can say with 100% certainty that drinking eight cups of water does not make a difference in your weight loss. However, if the results showed that drinking eight cups of water helped someone else lose more weight than it did for the participants who drank only the bottle of water, we would have to change our conclusion.
Because even though we didn’t find any significant effects for water on weight loss, other studies have shown that drinking eight cups of water helps maintain normal blood pressure and glucose levels, both of which contribute to overall health.
Ethical considerations
As mentioned earlier, making assumptions about other people’s behaviors is never good science. When conducting research with human participants, there are several important ethical principles that must be considered.
One of the most fundamental is informed consent. All study participants should give you their permission to collect samples of their hair and/or blood for use in scientific experiments. This includes writing down who they speak with while giving them this information as well as what questions are asked during the experiment.
By requiring informed consent, researchers are obligated to tell participants all relevant information about the study. It also means ensuring that everyone understands what will happen next and how personal data will be handled.
As we have seen, due to the sensitive nature of some studies, it is essential to have adequate protections in place to ensure confidentiality. If needed, researchers can ask participants not to discuss the findings outside the group or even keep it confidential from others.
General rules apply when asking someone to participate in your research. Make sure the participant knows why they are being asked to take part and what benefits they may get out of it.
Determining results significance
The second major factor in determining if research is significant is how much importance researchers give to its findings. This step comes after the initial study has been conducted and the results are published.
The researcher(s) must determine whether or not their findings are important enough to make changes to what is done already. If they cannot find an answer, then the conclusion drawn from this new information will likely be that nothing needs changing!
By having doubts about the effectiveness of a treatment, people can sometimes avoid it and suffer from its side effects. By thinking that a certain approach doesn’t work, you may stop trying it and look for something else instead.
That isn’t good for your overall health and wellness, but it also means you won’t get the exact same result you might have with the test intervention. It also means that more people will potentially be exposed to the risk if it does not work.
Meaning there is greater chance that someone will develop symptoms of the disease and/or pass along the infection to others.
