As parents, we desperately want to help our kids. Understanding research can keep us from jumping on every bandwagon that comes along and getting disappointed.
I was given great advice when my daughter, Hannah, was diagnosed with autism. I was advised to learn what constitutes good solid research so that I would be able to make informed decisions about the best treatments.
I’ve met several parents who refuse to try anything more, because they’ve attempted so many treatments that were not effective. They simply couldn’t handle the emotional trauma of another failed program. My heart breaks for them.
This is why understanding research is vital. We may not enjoy reading scientific information, but checking the research behind any treatment can save a lot of heartache.
Warning! Scientific information ahead! Don’t worry, I’ll try to put it in English.
Let’s look at a hypothetical situation with you as the researcher. Let’s say you want to test a new drug, which is supposed to increase concentration, as a treatment for autism. You first conduct a thorough search of resource books, journal articles, etc. to find out what is already known about the drug. Look at information that supports the effect of the drug, as well as information that does not support it. From this, you develop a hypothesis, an educated guess, as to whether or not the drug will be effective.
You have a group (population) of 300 children with autism whose parents have given permission to participate in the research study. You want to be sure these children are as similar as possible. This way, if improvement is made in the children who take the drug, it wouldn’t be likely to have been caused by something else, for example a difference in age.
So you choose only the ones that are a certain age, say ten years old. Now since autism is a spectrum disorder with some children much higher functioning than others, you will also want to choose the ones that are as developmentally similar as possible. This can be done in various ways. You choose to have them tested for an overall developmental age as well as administering an IQ test.
Once all the testing is done, you have 100 ten-year-old children with very similar developmental ages and IQ’s. This is not the entire population – it is your sample.
You would divide the children into two groups: an experimental group and a control group. You do this in such a way that each child has the same chance of being in the experimental group as he/she does of being in the control group. This is called random assignment. This is done to further ensure that the two groups are as similar as possible.
Experimental Group and Control Group
The experimental group would receive the new drug. The control group would receive a placebo (a pill that does not contain any medicine). You decide that the children will receive the drug or placebo every day for ninety days. At that time, they will each be tested again as to their developmental age and IQ. You will then note the differences in each child and analyze that information.
Ninety days have now passed, the children have been tested, and you are looking at the differences in the control group. You find they each improved a little bit in their developmental age, but there was no improvement in their IQ’s. When you look at the experimental group, however, you find they’ve improved 6 – 8 months in developmental age and 4 -6 points in IQ. Why is this? Can you say that you have proved this drug works well as a treatment for children with autism?
Well, this is where it gets tricky. First of all, research does not “prove” anything. It can increase or decrease the certainty that something is probably true. There are many factors that must be considered. For example, there are variables that could have had an effect. Some examples are the following:
- the children may have gone to different schools could have used varying curriculum in mainstream classrooms as well as special education.
- The children may have had different teachers with different levels of skills and experience.
- Some of the children could have started other treatments for autism during the time the study was conducted.
There are also several types of bias that can take place in research.
- Sampling Bias – when the sample is not representative of the population. In our case, our sample consisted of all ten year olds. The drug may have a different effect on children of other ages. Our sample also started at a certain developmental age level. It could be that the drug wouldn’t be as effective on children of other levels.
- Placebo Effect – this is a bias that takes place when participants believe they are receiving the actual drug, when they are in fact, receiving the placebo. To lessen this effect, a single-blind study can be done, which means the participants don’t know whether they are receiving the actual drug or not.
- Experimenter Bias – when the researcher’s expectations have an effect on the results of the study. To alleviate this effect, a double-blind study can be done which means that neither the participants nor the researcher knows which participants get the actual drug.
Variables and bias could have had an effect on the outcome of the research study. This is why you cannot say with 100% certainty that the drug caused the change in the experimental group.
However, there are other things that can be done to increase the certainty, such as replicating the study and getting the same results. Or a longitudinal study can be conducted by testing the same children in the future to see if they maintained their gains in developmental age and IQ.
Understanding research is vitally important for your child’s sake and yours as well. Do not just assume that a treatment does what it is supposed to do. Always look at the research behind it.
Note: This page does not give exhaustive information on research. It only serves to give a basic understanding of experimental research. It does not discuss ethics, validity/reliability, or data analysis, etc.