Explain how you would design a research study to compare the effectiveness of two weed killers. Both weed killers A and B are known to be effective but it is not known which is more effective and on what type of weeds. Weeds can be broadly classified into “broad leaf” and “herbaceous” types.
a) Name the explanatory and response variables in this study.
b) Identify a confounding variable and state the type of experiment needed to account for it.
c) Outline an appropriate design, clearly describing all the important features needed in any well-designed experiment. An annotated flow-chart may be used.
And what if i'm having problems cause i did not attend class for these?
Anyways From what i can figure out:
a) The explanatory variables would be the actual weed killers (A and B) and the different type of weeds themselves... so two explanatory variables ?.. weird?
Response variable .... not to sure... but wouldn't it be the effectiveness of the weed killers?
b)A confounding variable..... Would that be something like weather when it is tested? not really sure about this...
c)experimental design with control/randomization/replication?
(do we just use one)
maybe if we use control then we can use comparison of weed killers and compare there effectiveness?
PLease need some help with this.
Thanks
Glad to see some of your thinking.
I think what you call "explanatory" and "response" variables are what we call independent and dependent variables. "Confounding" = extraneous. Although weather is an extraneous variable, weather effects can be controlled by using a greenhouse.
I think what you have here are two separate experiments, one for “broad leaf” and one for “herbaceous” types.
Possibly this can help:
An independent variable is the potential stimulus or cause, usually directly manipulated by the experimenter, so it could also be called a manipulative variable.
A dependent variable is the response or measure of results.
Extraneous variables — other than the independent variable — potentially can affect the dependent variable, so they must be controlled. If possible, you try to keep them constant between the experimental and control group.
The experimental group receives the independent variable.
The control group is similar to experimental, except it does not receive the independent variable. Extraneous variables are balanced between experimental and control groups.
Types of experiments
1. Single blind gives the control group a placebo — a substance that is inert, it has no physical effect. Subjects don't know if they are in experimental or control group to reduce placebo effect, a change in behavior solely due to believing that you are getting the independent variable.
2. Double blind keeps both subjects and experimenter ignorant of group setup. Distribution of the independent variable and placebo are controlled by third party. This controls for experimenter bias and self-fulfilling prophecy, which means that experimenters with particular expectations are likely to consciously or unconsciously to bias the experiment and influence it to conform to their expectations.
As an example, suppose you want to find out if fluorides reduce dental cavities. You would find two groups, trying to control the extraneous variables. Extraneous variables are found by surveying previous research in the area. In this case, you would match the groups in terms of previous history of cavities, diet and dental hygiene habits including how and how often they brush their teeth.
The experimental group would get toothpaste with the independent variable, the fluoride, while the control group would not have the fluoride in their toothpaste. The toothpaste without the fluoride would be the placebo.
The dependent variable would be the number of cavities after participating in the experiment for a time. The dependent variable indicates the results, but it is not the results. At the end of the experiment, both groups could have no change in cavities or one of the groups could have a greater reduction in cavities. (Of course, if the fluoride increased cavities, you wouldn't want to use it.) All of these varied results would be indicated in terms of the dependent variable.
If only the subjects do not know who is getting the fluoride, it is a single blind experiment. If both the subjects and experimenter do not know, it is a double blind.
I hope this helps.
It did help but not sure if they are two different experiments cause your are comparing the effectiveness of each weed killer.
Also not really sure on how to draw the flow chart for the design.
a) The explanatory variable in this study would be the type of weed killer used (A or B), and the response variable would be the effectiveness of the weed killer on the types of weeds (broadleaf or herbaceous).
b) One potential confounding variable in this study could be the initial weed density. If one treatment group starts with a higher weed density than the other, it could affect the effectiveness of the weed killer. To account for this confounding variable, a randomized complete block design (RCBD) would be needed.
c) An appropriate design for this study would be a randomized complete block design (RCBD) with two treatment groups (A and B) and two types of weeds (broadleaf and herbaceous).
Here is an annotated flow-chart explaining the design:
1. Randomly assign the experimental units (plots) to one of the two treatment groups (A and B). This ensures that the assignment is unbiased and minimizes the impact of any potential confounding variables.
2. Divide the experimental units into blocks based on the types of weeds (broadleaf and herbaceous). Ideally, each block would contain an equal number of experimental units from each treatment group.
3. Within each block, apply the assigned weed killer treatment (either A or B) to the respective plots. Ensure that the application is done uniformly and according to the recommended application rate.
4. Measure and record the effectiveness of each weed killer on the different types of weeds by assessing the weed reduction or elimination after a specific time period. This could be measured using metrics such as weed coverage or weed count.
5. Analyze the data using appropriate statistical methods, such as analysis of variance (ANOVA), to compare the effectiveness of the two weed killers on the different types of weeds. This analysis will determine if there are significant differences in effectiveness between the treatments.
By using a randomized complete block design, we can control for the potential confounding variable of initial weed density. The blocks ensure that the different types of weeds are evenly distributed between the treatment groups, minimizing the impact of weed density variations on the results. Randomization within each block further ensures unbiased assignment of the treatments to the experimental units.
Overall, this design allows for a fair comparison between the two weed killers (A and B) and provides insights into their relative effectiveness on different types of weeds.