Bio Analysis Test 1 Forum Questions
20. Distinguish between basic and applied science. Is one "better" than the other? Explain.
Basic science is understanding fundamental problems. Such as what makes up wood and bark. Applied science is aimed to solve practical problems. Uses the knowledge gained from basic science and attempts to solve practical problems. Such as which wood would be better to build a table. Neither should be considered "better" since both work together. It is difficult to solve a practical problem without the fundamentals. Codependent
5. Why can't a scientist work in isolation?
Because science depends on interactions within the scientific community. It is best to have 2nd, 3rd, and 4th opinions to encompass more of the subject at hand and provide validity Collaboration is important because sometimes, an individual needs the help of another to further understand their process.
3. Why isn't science predictable and predetermined?
Because science has the tendency to be ever-changing. Different processes are being carried out by different individuals at different times. Our knowledge of science increases with each passing year due to increasing resources and technology.
10. What is data analysis? Why is it necessary in science?
Data analysis allows scientists to guage whether or not the data supports the hypothesis or opposes it. Also it can prove that more data might be needed. It is necessary in science because, without data analysis, scientists could assume incorrect conclusions and could mislead a significant amount of people
18. Define/distinguish between empirical and deductive science. Explain the strengths and limitations of each.
Empirical science is systematic and uses observation, testing, and repetition to find understanding. It is also measurable. This useful because it can provide clear answers. However, the weaknesses with this type of science are that it assumes that the natural world will always and have always operated in the same way which is not necessarily true. Deductive sciences is based on models. A small scale model could be used to make some type of prediction about a real life event. This is useful because it allows a certain event to be tested with smaller costs, and with, more likely, less time. However, the downside is that a model will sometimes have a very hard time replicating the real life event--even with multiple changes.
18. What do you learn by each of the following: exact replication, systemic replication, and conceptual replication?
Exact replication - you could learn mistakes in the original experiment and correct them and see if the results would change. Systematic replication - by changing something in the original experiment, you could find that their means for finding an answer were too unspecific. Maybe, the population the original experiment tested was too small, or was not spread out enough. Conceptual replication - you could find out if the specific way the experiment was set up was compromising the validity of the results. For example, testing the a car's breaking ability. The original experiment tested distance, but our new experiment tested time.
19. Define/distinguish between experimental and historical science. Explain the strengths and limitations of each.
Experimental science is based on direct experimentation in present times. This is useful because data can be calculated in real time and it can be repeated. However, some experiments could prove too large or too small to test. ALso, does not attempt to explain anomaly events that may have occurred in the past that seem to defy the natural laws. Historical Science is trying to explain what has happened in the past using today's evidence. A strength of historical science is that there is no need to retest the phenomenon multiple times. However, today's information can often prove to be incomplete and will lack key parts. Scientists are unable to know the exact conditions in the past and therefore experimenting and formulating an explanation as to why the event took place is not as plausible as simply describing the events that took place. Such a branch of historical science consequently fails to meet the demarcation argument that it must be directly observable and testable.
12. What makes scientific idea more trustworthy and more likely to be true?
Experimentation, if applicable, will yield very similar or even identical results almost every time. Also those results should be peer-reviewed by other scientists that are experts in that field. Also that experiment should be repeated multiple times and with different types of people
9) The results of an experiment in which teaching method is the experimental treatment, used with a class of low achievers, do not generalize to heterogeneous ability students. What type of threat to validity would this be? Explain.
External validity because the experiment is very specific which makes it hard to generalize to other situations.
1. What is a bivariate graph?
Graph with two variables. The dependent variable is usually on the y-axis, while the independent variable on the x-axis. All tick marks on the graph are all usually not labeled (every other most of the time - or else too crammed).
20. Give an example of a 2 x 4 factorial design in an experiment. Explain why it is factorial and why it is 2 x 4.
If I planted some flowering plants into a pot of soil containing a growth serum. There were two types of pots with two different serum concentrations and I measured them at 4 different time intervals--let's say once a week for 4 weeks. It's a 2x4 factorial design because there are 2 independent variables and there are 2 types of one and 4 types of the other. It is factorial because there are multiple factors affecting each subject at once.
1. "The behavior of the researcher may influence the independent variable, thus threatening the validity of the study." Give an example identifying a possible behavior that is a threat to validity.
If a researcher was hired under a company concerned with cleaning detergent. His behavior in the study could be affected by his motivation. His goal is to present how their brand is the best with research. However, the independent variable, the type of detergents used, could be greatly skewed. The researcher could compare their brand to other "not-so-good" brands as to their direct competition. Also, they could magnify the results and stretch their wording and make is seem like the difference was greater than it actually was. Thus, threatening the validity of their study.
8. What is required to come up with a valid conclusion?
If it is warranted by a significant amount of evidence and testing. However, conclusions can be revisable. Careful deductive and inductive processes must be used to run multiple experiments which will result in evidence that is either sufficient enough to support the hypothesis or that does not support the current hypothesis.
12. When would a subjective measure be preferable to an objective measure?
If multiple subjects all performed on a very high standard, and it is hard to tell which one actually did better because they all surpassed a certain measurement of "skill." Making their performance subject to a knowledgeable researcher could make the results more clear.
16. Why is scientific replication important?
If research can be replicated and tested more than once, it provides more clarity and reliability. Replication can allow the original designer to see their mistakes and get a conclusion backed with stronger evidence. When the same answer comes out every time, then in the real world, it is more likely to be true every time.
11. What is the difference between sample size and the number of replicates in an experiment?
Sample size is how many total experimental units you used. Replication is the repeated application of the treatments to multiple independently assigned experimental units.
11. How do scientists use evidence to choose between multiple competing ideas?
Scientists most ideally chose the argument that contains the largest body of reliable evidence, aligns with the current body of evidence and has zero or the least amount of counter evidence. Evidence can be either direct of circumstantial. Either of the two can be used to back up any of the competing ideas. Ideally, the idea backed with the most evidence will prevail as truth.
6. What is it important to show variation in your graph?
Show the range, standard deviation, standard error, or confidence intervals within graph. This allows the reader to have a deeper understanding of the results. With this, the reader can see the data sets instead of the average of the results. With small variance, the reader will know that the results are a reliable representation of the results of an experiment, while in contrast a big variance will show the probability of how unlikely the result of an experiment is true.
9. Give an example of a situation where a stratified design would be desirable.
Stratified design is useful when you want to run an experiment on a larger population size. If you wanted to test on a certain variable, you could organize the population according to your variable (organize them into stratifications) and pick within those strata to come to a conclusion.
4) Identify the variables in the experiment below.
Independent variable: presence and concentrations of ATR and CPF, whether one at a time or in a combination. Dependent variable: mRNA levels of DMNTs and MBD2s. Confounding variables: brand/manufacturer of ATR and CPE, the type of carp, time of day of treatment, season the of the 40-d incubation period, size and age of carp.
2) Distinguish between the independent and the dependent variable in an experiment. Give an example.
Independent variables are the variables being changed. Dependent variables are what changes because the independent variable was changed. An independent variable could be sunlight and the dependent variable could be the turgidity of a plant.
3) Identify the variables in the experiment below; all the variables.
Independent variables: concentrations of atrazine and fenetrothion NOEL Dependent variables: the binding of corticosterone (B) to CBG in an amphibian and a mammal. Confounding variables: type of plasma, the amphibian used, the mammal used, the number of amphibians or mammals, the same brand of atrazine and fenetrothion NOEL, performing the microdialysis protein assay the same way each time, using the same volumes throughout each sample, using the same size and age toads or rats
7) List and briefly describe the different types validity.
Internal validity is identifying all confounding variables and making sure no area of information was missed during experimentation. Causality must be established. The independent variable must be responsible for the changes seen in the dependent variable. External validity - the results must be generalized enough to be applicable to more generalized scenarios found in the world. Construct validity is based on the ability of the experiment to address the question. An experiment could be done, but with no helpful conclusion at the end of it all. Statistical validity is whether or not the proper statistical analysis was used. For example, whether or not a chi-square would be the proper test
8) I set up a blind to observe the behavior of lions. What type validity might this threaten? Explain.
Internal validity: the lion may notice the blind which may cause it to behave strangely.
2. What can graphs tell you about a data set that simple descriptive statistics cannot?
It can show us the significance of the relationship between variables, and can reveal a qualitative pattern while statistics cannot. Graphs are unique because depending on how they are created, they can either reveal or hide the meaning of the data set.
5. What is a placebo? Why is it needed? Explain.
- a fake drug - helps isolate the results without the drug actually having any significant impact - needed so we are able to see if there is a shift in results just because the subjects "knew" what they were getting. For example, if a certain drug is supposed to heighten dopamine levels, but the placebo was given, and the subject said they did feel more on edge or have heightened senses. The subject says they do feel it just because they knew what they were getting. Because of they think it, there is an actual slight increase in dopamine levels without the use of the real drug. Now, with a subject who took the real drug, we can analyze how much their levels went up, compare it to where the placebo analysis was and have a clearer view of it's actual effects.
23. What happens when the sample size is too small?
- could get inaccurate results b/c there are always outliers and random exceptions. - outliers in a small sample size can have a much greater effect on data and skew your findings drastically.
12. Critique the attached graph. Describe at least two things that could be improved.
3D aspect of the graph is misleading and makes it hard to comprehend the data. I find it unnecessary for there to be percentages next to the pie graph sections. (Put percentages in a table?) It does not follow Tufte's Rule.
11) A researcher measures a distance at 1.315 m The actual value is 1.200 m. What is the percent error?
9.583%
15) What is the difference between a correlational study and an experimental study?
A correlational study can be conducted in natural or artificial settings where a relationship is found between two preselected variables. Have a lower level of constraint and do not establish causality. An experimental study has a higher level of constraint because confounding variables are being controlled for. Due to specific results where the independent variable is the sole influencer of the dependent variable, causality may be established.
16) What are strengths and weaknesses to having a high level of constraint?
A high level of constraint allows one to be very specific in their findings, conduct a detailed procedure, and write complex analysis, but at a cost of applicability and realism. Although it allows one to be in more control, it can take more time and money. Also, it gives the researcher assurance that the findings are not due to confounding variables. The more constraint put upon an experiment causes the findings to be more specific to that setting which may not be the most realistic.
14) What does it mean when we say that an investigation has a high level of constraint?
A high level of constraint means that there are a lot of confounding variables being accounted for and controlled.
8. What is the difference between a histogram and a bar chart?
A histogram shows continuous data (distribution of a given value in intervals) while bar charts show categorial data (comparison between data).
17. How is a hypothesis different than a theory? How are they similar?
A hypothesis is a suggested explanation for an action or some type of phenomenon. A theory has been tested and has a unified explanation.
12) Give an example situation where a researcher's experimental design does not show good construct validity.
A researcher might want to use how much you enjoy your job to explain a person's general happiness. However, an individual's happiness isn't solely due to their job. For example, someone could be enthusiastic and happy with their job but could be in an abusive relationship and have family problems outside of work.
11) Give an example situation where a researcher's experimental design does not show good internal validity.
A researcher wants to improve students grades in a class so he concocts a genius idea to bribe them by giving them treats every time they do well on a test. THis is an example of bad internal validity the researcher isnt able to conclude that the candy was the only variable that caused the students to do well in the class. There may have been other reasons such as personality, competition, or motivation.
3. Give an example of a single blind study. Why was the single blind study used?
A simple single-blind experiment involving two random groups and neither group knows whether they are receiving a placebo drug or the actual drug that tests alertness, however, the researcher knows which group received the placebo or the drug. The single-blind study was used in this scenario to prevent the participants from reacting a certain way based on the drug they have taken and provide a more accurate, unbiased analysis. If the participants knew that the drug is testing for alertness, they may start acting more alert due to the absence of a blind.
4. Give an example of a double blind study. Why was the double blind study used as compared to a single blind study?
A taste test would be a good example of a double bind study. The double bind study would be more useful compared to a single blind study because it eliminates bias on two levels. In a single bind study, if the researchers already had their own preconceived thoughts on what would taste better or worse, then their placement of information or their analysis would already by compromised.
7. What is a contour plot?
A technique by which one can show results in 3D on a flat surface without visual trickery. There is a plane for z values within the x and y coordinate graph.
8) Distinguish between accuracy and precision
Accuracy is how close you are to the target, precision is how close your shots are to each other.
21. What are advantages and disadvantages of the repeated measurement design?
Advantages of taking repeated measures - less likely to record mistakes - Double checking, triple checking, and even checking 4 times can make sure your results are accurate Disadvantage -if the problem resides in the subject, then each measure was pointless to begin with - it takes up for time, and depending on the experiment that could mean years.
21. What is an empirical observation? What makes it empirical?
An empirical observation is made in real time by the observer. It is made empirical because the human senses are used to assess the information.
19. Give an example of a nest treatment design. Explain what makes it nested.
An example would be to measure the strength of a power lifter. You can only analyze the strength of the lifter by themselves and their own ability. You cannot analyze their strength by measuring and accounting for their interactions with other power lifters. These power lifters were also separated in different groups within the experiment. For example ages 15-20, 21-25, and 26-30. Within each group, we could measure the degree of their strength giving a hierarchy or results.
9. What makes pie charts easy to misread?
These are used for visualization, and many times can be hard to understand which can lead to false conclusions. It does not easily compare values. An example is how the pie chart only shows percentage, but we do not know what the sample size of the experiment.
22. What implications does the following statement have for science? "We make our observations through the filters of our world view."
This means that our preconceived beliefs could influence the way we see our results, this leading to inaccurate conclusions. Seeing a certain conclusion through our world view could make a correct answer seen incorrect to us. It is possible for two individuals to interpret the same set of data completely differently.
13) I wanted to study the mating behavior of a species of rock iguana in their natural habitat. What are variables I would need to control for?
Time of day, the temperature, humidity, that you are watching iguanas that are actually ready for mating season, that you are watching iguanas all of the same species, etc
1) Write a good hypothesis and justify why it is good.
Under identical conditions, for a total of one month every night, full-time working adults aged 25-30 are less likely to move during their sleep when on a firmer memory foam bed compared to sleeping on a pillow top. This is a good hypothesis because the group is specific. Also, the number of turns and movements can be counted, thus being measurable. The sense of direction is firmness and the variables are clear. Type of mattress is the independent, and number of turns in dependent.
6) What does the "validity" of an experiment refer to?
Validity refers to the degree that the conclusion is supported by the experiment. For example, if the experiment is well-directed, highly controlled for confounding variables, and carried out using proper technique the resulting conclusion is likely to have more validity.
16. How could we control for edge-effect? Give a specific example.
We could control for the edge-effect by rotating the sample placements. For example, if there was an experiment concerning plants in a greenhouse, and the heater was placed on one side, then certain plants would receive more heat during the experiment. However, by rotating the plants' placement, we could help ensure an equal amount of heat transfer, ultimately leading to more accurate results.
9) How can accuracy and precision of a measurement be increased?
We could increase accuracy by increasing more trials and eventually calibrating our balance to fit the target number better. We could increase precision by decreasing the increments in your measuring device.
4) Explain why we replicate experiments.
We replicate experiments to make sure our results are a better representation of what we would constitute as "true" information. Essentially, it allows the variability of an experiment and the phenomenon in question to be estimated with more precision.
13. What does it mean when it says that the results were confounded?
When a result is confounded, it means that another variable, other than the independent, affected the outcome of the results in the experiment.
10) Nominal data must be mutually exclusive and exhaustive. Explain and give an example to illustrate.
When it says that nominal data must be mutually exclusive, that means that all the categories cannot bleed over each other, they are completely separated. Exhaustive means that all subjects meet one criteria, no singular subject is left behind. There is enough categories for all the observations. An example to illustrate this would be would be dividing a group of cars by color. There were 20 cars and we split them all by red, blue, white, silver, etc.
5. Give an example of when it might be useful to use a reference line or reference area in a graph?
When the author is trying to refer to a target or necessary goal for an experiment. For example, when there is a comparison between test group results and control group results.
7. What is a scientific argument?
When two or more scientists engage in a debate about a scientific explanation and have two different opinions. They use claims and empirical science to back their side of the argument. The statement may be formed as a compilation of scientific idea, expectations and observations which produces a guide for scientists in their research
24. Can the sample size be too large? What happens if the sample size is too large?
Yes. If it is too large, it can lead to unnecessary time spent on the experiment, with insignificant differences
25. Do you think that the term scientific creationism is an oxymoron? Why or why not?
Yes. There is no actual science behind a creationists position. Creation is based on faith in a greater being which has no actual evidence backing it up. Science is evidence-based where as creation is based on faith or religion.
8. What do you learn from subsampling?
You can learn the consistencies in your experiment when you subsample
3) Read the following scenario and answer the questions A researcher conducts a survey over time of a certain species of mice at different elevations. She captures ten mice at an elevation of 250', ten at 1000', and ten at 10,000' with all mice were collected on the same day. This procedure is followed once a month for three different consecutive months. Each mouse was massed three times. She determined the average mass, with variance, of mice captured at each elevation for a single day. She then compared the average mass of mice at each elevation on each of the three months. Then she pooled the data by elevation and compared differences in average/mean mass at each elevation. a. Why capture ten mice at each site instead of just one? b. Why mass each mouse three times? What do I learn by doing this? b. Why calculate the variance of the mean/average? c. Are there any possible confounding variables? If so, how can I control for them? d. Are there any threats to validity? e. How many times was the experiment replicated? Explain and justify your answer. f. What is the sample size? g. Was there any subsampling in this investigation? Explain h. and justify your answer.
a. To increase sample size and to reduce the chance or random error making a larger difference and significance on her study. b. Massing three times will give a better idea of the mouse's true weight. When you measure three times, you might see a fluctuation in wait, though very small, but will allow you to have better control of the confounding variables. c. Calculating the variance and mean/average allows a researcher to get as close to true value as possible and make it clear how precise their measurements were. There is no one true weight for the mice at each elevation, so getting an average gives us an idea and a number we can analyze. d. confounding variables could include the gender of the mice and their weight. Are any of the mice pregnant? You could control for this by getting more than 10 samples per elevation each time and picking 10 that all look the same and seeing if you can get a 1:1 ratio of females to males. e. There is some construct validity problems. What question was she trying to answer anyways? And is taking the mass the actual answer to what she was trying to find? f. I would say that there were no replications during the experiment because she pooled the data all together. g. 90 h. Yes, there was subsampling because each group of 10 mice was pulled from the same area at the same time.
2. What is the role of creativity in science?
it allows for an avenue to create new questions and ideas and see things from different perspectives. To further these ideas, we require creativity to produce ways to prove or suggest that these ideas are useful or accurate in the scientific world. Science does not have a guide book that tells us how things go; there are many routes a scientist could take to solve one problem, and the source of all these routes is creativity.
1. Why is science difficult to define?
"Science" attempts to cover and explain every phenomenon that occurs in the world; however, the ways of approaching different branches of science can conflict with one another which makes finding a core definition challenging. -covers many topics and ideas.
14. What methods are used to reduce error variance? Be specific.
- Increase the level of constriction - double check the quality of the items being used during experimentation, - make sure you have a large sample size and select randomly.
6. What are some past technologies that have facilitated large scientific advances?
1. Electron microscopes: allowed for the examination of small shit (bacteria) 2. Electrophoresis 3. the light bulb 4. The automobile 5. incubator
1) I was determining how lethal a new pesticide is to fruit flies. I cultured a bunch of fruit flies in vials and then divided them into six groups. Three groups received the pesticide and thee groups did not. 1. Do I really need three control groups? Explain. 2. Analyze this experiment for replication. What is the sample size (n = ?)?
1. I don't think that you really needed three control groups since the goal is too see the effects of no change. Why not just keep it to one group and just increase the sample size? It would be easier to analyze and would probably give a better reality of a control since fruit flies vary so much in their phenotypes. However, I would not say that having 3 control groups is generally a bad idea. 2. The sample size would be 3 if the other 3 were deemed control groups.
4. What are the graphical cues ranked from most to least accurate graphical perception?
1. Position along axis 2. Length 3. Angle or Slope 4. Area 5. Volume 6. Color or shade
4. How is knowledge obtained outside of science?
1. Tenacity/Tradition - believing something to be true because it has been around for a long time. 2. Intuition - believing something because it feels right. "gut feeling" 3. Authority - accepting ideas because they come from a higher respected authority. Rationalism and empiricism is what makes up science.
3. Summarize the general principles to consider when constructing a bivariate graph?
1. There are only two variables - one dependent on Y-axis, and one independent on X-axis. 2. The lettering is large enough to be read so even when the graph is shrunk, it will still be readable. 3. Make sure all the symbols are used within the graph, and is differentiable (open circle and closed circle). Make sure that there is a figure caption below the graph. 4. Make sure the axis is appropriate for the data and does not have empty space. (X and Y data has minimal values so it is appropriate) 5. Make sure there are no extra lines within the graph. 6. If definitions are too long, include them in the caption. Do not make the graph too hard to understand. Keep it simple. 7. Lines can be used to connect data points to emphasize a trend. 8. Origin does not have to always be 0.
2. Describe four things that can be done to keep the behavior of the researcher from threatening the validity of the study.
1. To make sure that the motivation for the study is unbiased and the search for the answer is not really directed in a preconceived path. 2. Make sure the researcher is knowledgable about the subject being researched. 3. Don't let expectations make you see the results in a different light. 4. use different types of blinding techniques.
13. Why are assumptions necessary in science?
Assumptions are necessary in science because they provide a starting point. Without assumptions, there would be nothing to test. Leads to the development of important experimentations and discoveries.
10) In a physical performance experiment, the pre-test clues the subjects to respond in a certain way to the experimental treatment that would not be the case if there were no pre-test. What type of threat to validity would this be? Explain
Construct validity because the way the experiment is set up, the results would be flawed since the knowledge of the pre-test would have swayed the results. Because of the way the experiment was constructed, the validity of the results would be questionable.
5) Why isn't high correlation enough to establish causation?
Correlation is the linear relationship between two variables which means that as one variable falls/rises, the other variable falls/rises as well. However, this mere coincidence that the two variables fall together or rise together cannot conclude that one caused the other. It may be due to something completely unrelated or by random chance. As talked about in class, grades and the amount of caffeine consumption have similar relationships on a graph but are not related, it is due to coincidence.
6) What is count/enumeration data? Give an example.
Count data is data than is both nominal and discrete. Count data could be given as 6 birds, 3 cats, 10 dogs, and 29 fish.
6. Distinguish between factorial, nested, and gradient designs.
Factorial treatments - you can determine the interactions. For example, you can place the actions with others on a measurement. Nested treatments - you can only analyze within subjects, as in you can only follow what the subject is doing on hand within their own atmosphere, as compared to what they do with others. Gradient treatments - This is used when direct manipulation is harder and the variable treated is continuous. This is when you cannot really measure the actions perceived.
17. Define love. Give an operational definition of love - one that can be used to make scientific observations
Love defined would be providing the highest good to the other. This could be seen in an operational definition as watching a mother animal feed its children first before she feeds herself.
10. What is Tufte's Rule?
Minimizing the ratio of ink to data. It maximizes the dimensions of information while it minimizes dimensions of presentation.
9. Does science only work by means of experimentation? Why or why not?
No. The reason is because there are certain theories we cannot test and experiment directly. For example, there is historical science; we cannot run an experiment based in that past that is unrepeatable. Another example includes subatomic particles (too small). Sometimes, consistent observations can result in accepted facts. For example, there is a general consensus that the earth is round. While it is hard to conduct experiments to "prove" it, many observations have been made to verify this fact.
10. Distinguish between objective and subjective measures. Give an example of each.
Objective measures - more focused on if you did or didn't do it - there is no bias or prejudice about it. An example would be if a insect died or lived during an experiment. A researcher cannot have a bias or prejudice about whether the insect died. Subjective measure - allows a measurement to take place and that measurement based on interpretation An example could be the results of a new dish washing soap. The researcher can wash plates with identical "dirt" and the effectiveness of the detergents would be subject to his discretion and could be measured on a scale based on effectiveness.
18) Give an example of a type of investigation what would have a low level of constraint.
Observing how baby pelicans get their food in the wild.
15. What is peer review? Why is in useful? What are potential problems?
Peer-review is when other experts in the field critic and help tune others work. It is useful because it provides another set of eyes to evaluate your work, potentially reveal flaws that you may have not considered. It can be harmful because it may cause unintentional plagiarism.
2) Discuss what determines the number of times an experiment should be replicated?
Power and effect size should determine the number of time you should replicate an experiment. If you replicate a certain large number of times, you may be more likely to see a difference (power). But, the more times you repeat that experiment, the significance of that difference (effect size) may also decrease.
7. What is the difference between statistical power and effect size?
Power: how likely you are to detect a significant difference if one exists. Effect size: how meaningful is the difference.
22. What are pseudoreplicates? Give an example.
Pseudoreplicates are when you repeat an experiment, but the experiments are not totally independent of each other. For example, you test certain pesticide levels in a lake nearby a farm. However, your samples were all pooled from the same lake and within very close times of each other. This also includes using subsampling as replicates.
14. Why is publishing critical to progress in science?
Publishing is critical because it allows other scientists to evaluate your research and potentially stem new ideas. It's important to hear other perspectives to refine your research. This allows other to retest and come up with conclusions of their own and contribute to a greater understanding. Also, when different minds see the same information, it allows that idea to advance faster.
5) Give an example of the use of randomization in an experiment. Be specific.
Randomization in an experiment could be put into example when conducting an experiment about the effects of ambient music during studying. We could randomize our subjects and randomly pick 100 people aged 18-22 and ask them to take two tests based on similar information and compare their scores with and without music in the background.
15. What is randomization to and experiment and why is it important?
Randomization is when you pick from a selected population in order to run tests, and eventually there will be a relatively even amount of differing variables. It is important because it controls for an unknown bias. A researcher could intentionally pick his subjects, thus leading to invalid results.
24. How do Rationalism and Empiricism work together?
Rationalism and Empiricism, together, make up science. Empiricism can handle the measurable, and rationalism the unmeasurable. Rationalism can help reason out the results seen by our senses and form a concrete conclusions ready for publication. They can test experiments and theories, in a check and balance system. Empiricism utilizes measurements and observations, while rationalism utilizes reasoning and premises from empirical evidence to come to a conclusions
23. Distinguish between Rationalism and Empiricism. How does each "work?" What are the pros and cons of each?
Rationalism is thinking in which knowledge is developed through reasoning and logic. Rationalism is knowledge that can be gained without experiencing it directly but by using reasoning. A pro of this method is the lack of measurement, which could save a lot of time. However, we can make great leaps in knowledge through intuition without backing those theories up which is no bueno. Empiricism is gaining knowledge through observation with our own senses. A pro of this method is that it is hard to deny what happens right before us, compared to historical science. However, our senses can fool us and it can be hard to believe abstract concepts which a human cannot physically interact with.
17) Give an example of a type of investigation that would have a high level of constraint.
Testing the effects of a new drug that fights a fatal disease.
13. Critique the attached graph. Describe at least two things that could be improved.
The 3D aspects of the graph makes it hard to comprehend the data. There are no units within the graph. The variables are not easily differentiated. Add axis title.
11. What about the attached graph is misleading?
The first is the dates of the growing bar graph and the bar graph on the bottom that is not filled. For example the year 77 on the top shows that is goes from the year 76-77. However on the bottom it only says 77, which is very confusing because the reader does not know if it is 76-77 or 77-78. Also the graphs on the tops show a trend of increasing growth, but the graph on the bottom does not show the same pattern. This is very confusing to the reader because there is not consistency with the data. there is no explanation or reason why there is a division into two parts. The 3D aspect of the graph is misleading, which makes the data hard to comprehend. It also does not tell us how much total aid (exactly) is given to localities.
7) Classify each type of data. Justify your answer. a. 23.4 m b. 45.6 girls c. Ford d. 24 freshmen, 35 sophomores, 45 juniors, 1 senior e. Seven f. 2 agree, 3 disagree g. 35 km/h h. Class two tornado i. 2 juveniles (2-12 cm), 4 adults (12-20 cm) j. The number of wins and losses during the basketball season
a. this is coninuous data because it can be represented in a decimal as well as having a unit. Also, there is no real limit to the measurement. b. nominal and continuous because girls implies a name, and continuous because it is not a whole integer. c. nominal because there is a name d. discrete and ordinal because we have whole numbers and the nominal names have a ranking, thus making it ordinal. e. either discrete because its a whole number, but it could also be continuous because you could continue in a decimal fashion. It depends on the circumstances. f. discrete and categorical; discrete because there are whole numbers. Categorical because each answer can only be one or the other and they are separated. g. this is continuous because it is a measurement with units. h. this is categorical, because the class 2 implies there are other classes that the tornado could fall into. i. this is categorical and discrete because the subjects can be placed into two different groups. It is also represented in whole numbers. j. This is categorical because there are two groups given, either a win or a loss.
