Designing an experiment plays a big role in the type of conclusions you can draw from your data. If you design an experiment in a different way than how you want to interpret your data, then your interpretation of your data would be paused or skewed. If you wanted to know how many M&M’s are in a jar, then you would design the experiment to calculate how many M&M’s are in the jar. This would allow you to draw the conclusion at the end of the experiment that the number you have quantitatively assessed is the number of M&M’s in that jar. If you started out your experiment wanting to know how many red M&Ms are in the jar and then changed your mind and decide to find the total number of M&M’s in the jar. This means that this experimental design for just red M&M’s would render only red M&M’s total amount in the jar rather than the total number of M&M’s.
An example of experimental design and drawing valid conclusions would be the roadkill presentation from BIOL 251. For this experiment, we designed how we would drive around and count how much roadkill we found within the Farmville area. We had to decide what type of roads we wanted to use and what kind of speed limit we wanted to look at. We decided we would look at a highway, 460 West, and more of a rural road within Farmville, Back Hampden Sydney. Using these two different types of roads we were able to make conclusions about the amount of roadkill in a more urbanized area compared to a rural area and also if speed limit played a part with the amount of roadkill. Another data point that we decided to look at was the weather so on our sheet where we counted how many roadkill there were we also had a column for what exactly the weather was like at that time. This is why I wanted to also draw conclusions on whether or not weather and the condition of the road whether it be wet from rain or cloudy if that caused increase or decrease in the amount of roadkill found on the road.
BIOL 251 Research Showcase Presentation 2019.pdf
In BIOL 350, we looked at the global extinction of reptiles. We made an experimental design by using the data that was collected by other organizations dealing with the extinction of reptiles. We created a mass list of the extinction of reptiles data and compiled it from the website. From there we designed how we were going to categorize each reptile mentioned within this list. We basically took the list and the data we found on the reptiles and created four graphs. These poor graphs were temporal patterns of extinctions, extinction casual factors, mainland versus island extinctions, and spatial distribution of extinctions. From creating our experimental design while we did not go out and do the data collection ourselves, we were able to compile the list from websites to be able to make these lists and draw conclusions about what exactly was happening to reptiles.
The immunodeficiency presentation project that we did in biology 404 also shows how an experimental design can slowly lead you to conclusions. This experiment started out with us receiving flow cytometry data analysis for a patient and then as the weeks went on in the lab, we received more data. This data came from us making conclusions about the previous weeks data. Using the previous week’s data to create a new “experimental design” for the next week allowed for us to be able to draw a conclusion about what the patient has. There was a mix of qualitative and quantitative data for this experiment. Data from the flow cytometry and other graphs could be considered qualitative, while some of the other data we’ve received was quantitative. Overall, this experiment allowed us to create an “experimental design” each week based off of the previous week and all of this data eventually LED us to the conclusion of what the patient has.
BIO 404_ Imminodeficieny Presentation .pdf
Whether an experimental design is doing an actual experiment or taking pre-existing data to complete a quantitative analysis of the data to draw a conclusion both types of experimental design is important. Sometimes there’s an abundance of pre-existing data but there was no data analysis on a specific part so taking the pre-existing data and just focusing on the data analysis is important. While there are other experimental designs that focus on gaining data so you’re able to do data analysis later in the experiment. Both types of experimental designs allow for valid conclusions that can be made after data analysis occurs.