The box plot is lazy bell curve!

Multivariate Analysis (ANOVA) / Mar. 8th

You need a Multivariate Analysis (MV) test to evaluate multiple comparisons. With five catagorical varying treatments/factors/levels to test, the result would be 15 tables using a t-test analysis. By using an MV test, there would only be one table to review.

Plus, every time that you apply a test, there is a chance that it is wrong. With 20 tests there are good odds that the statistical conclusions are a mistake, in at least one (a one in twenty chance).

ANOVA: Analysis of Variance looks for any statistical difference across groups. How much do the results deviate form the mean? (Note: variance is the square of the standard deviation). The "R" foundation for statistical computing (freeware) can analyze this to run the data, but you would drop the results into Excel to better visualize them.

The Tukey is always the post-hoc test after the answer is found. What this means is the you need to group similar results to more easily see where differences are.

Introductions / Feb 27

Introductions: from general to specific. What is missing? What is interesting to others? How can I fill in the gaps?

Introduction

Phenotypic plasticity, i.e. the ability of a particular genotype to produce different phenotypes in response to environmental variation (West-Eberhard, 1989; Thompson, 1991: Via et al., 1995: Zhivotovsky et al., 1996; DeWitt et al., 1998; Pigliucci, 2005), has been the object of considerable interest and debate over the past two decades (Brookes 2007). Will snails grow significantly different based on their location? There are undoubtedly a variety of conditions that would influence snail size, including current flows (Palmer, 1979), temperature (Atkinson, 1994: Palmer, 1979: Trussell, 2000), area predators (Brooks & Remy, 2007: DeWitt et al., 1998), food availability and type (Burrows & Hughes, 1991), optimal digesting (Menge, 1974: Burrows & Hughes, 1991), and genetic factors (Thompson, 1991: Marko, 1998: Maruyama and Birky, 1991). Another consideration would be to analyze the affluent conditions, or chemical composition to rule out area contaminants as an influencing factor.

Our analysis we will start with the baseline of having no altered conditions. With the use of existing empirical data (Price 2007) this will be reviewed. Our particular study looks at the species Nucella lamellosa, known more commonly as “Dogwhelk,” a snail that lives in the Pacific Northwest. Total mass measurements were taken of this species, in three samplings over a 24 day period. The two locations that Dogwhelks were taken from, include False Bay, and from the north side of Cantilever pier at Roche Harbor, both locations are unique intertidal habitats on San Juan Island, in Washington state (Price, 2007).

Treatments were fed in the same way, with barnacles from a third locality, Jackson Beach, also on San Juan Island. All the conditions, including flow rates, were the same across all replicates in the two treatments (Price, 2010, personal communication). Both of the localities are protected from waves and have do have crabs (Price, 2010, personal communication). Locations are approximately 18 km apart².

t-tests revisited / Feb. 24

A t-test is used to determine the significance of the difference between two sets of data. In the experimental design snail changes of length and lip increment were measured during increased emersion body temperature conditions. The first t-test recorded differences in fraction of total length added. The second t-test measured increase of lip increment.

The control treatment was at 12 degrees, with two subsequent emersions of 20, and then 28 degrees. There was not statistically significant changes in the first temperature increased treatment. However, in the second, at 28 degrees, there was statistically significant increases in calcification morphology of length and width.

All errors bars represent one SD, and an asterisk (*) indicates that the response treatment mean is significantly different from those of other treatments.

1. For this graph, the research hypothesis tested is:
An increase in emersed body temperature will affect Nucella ostrina morphology.
2. The statistical null hypothesis for the data in the graph is represented by the control temperature of 12 degrees, which is the baseline.
3. The variables presented in the graph are the categories of total length and lip increment.
4. The statistical null hypothesis is rejected as the asterick indicates a significant difference.
5. The research hypothesis is accepted as mean growth in snail mass was significantly higher in the second temperature treatment.
6. The control treatment is 12 degrees.
7. The experimental treatments are 20 and 28 degrees, as described by Yamane & Gilman in their 2009 published report.

Response to peer review / Feb. 23

Dear reviewer:


You are correct in desiring for a more elaborate reason as to why further studies are needed for understand snail body mass increases. I did mention that our tests had insufficient variation to determine a solid conclusion. Just looking a difference in close proximity locations may not be a good as a sole reason.

Chances are very likely that food and water composition are not the only factors for size alterations. Predatory attacks also are a large component to as to how a snail will or will not gain body mass. The body mass also needs to be looked at to determine whether it is the shell or the soft tissue that is gaining in mass. Shells increase for several reasons including abundance of food and lack of predations. Soft tissues increase due to abundance of food, and adaption to current flows where stronger muscle attachment is a consideration.

Should you find an studies that relate further to current flow issues, do let me know. Thank you.

Part II response / Feb. 22

The three people who reviewed my paper seemed to feel that points were clear, although the discussion needed more development. With the option of reflection of what other potential issues might be for snail total mass changes, I explored other texts that had tested variables that could affect snail mass. This helped for me to articulate the basis thoughts that I had about this.

Although it seems like there is excessive citing, all were from authors who had performed experiments that were helpful to draw upon in conclusion. Four more citing were used from prior experiments to substantiate my views.

Revised t-test caption per Price request

Table 1. Nucella lamellosa.

Differences in total body mass by the gastropod for 3 different treatments,

in a submerged water temperature of 13°C for 24 days.

All error bars represent 1 SD.

Scale measurements/ Feb. 22

Variables are characteristics of a group of objects or events that can be measured over a number of different numerical values.

In science only one variable is often changed. There is an independent and a dependent variable. If there were no change, there wouldn't be a response by the dependent variable.

Continuous - is quantitative / Categorical - is not quantitative

Discuss how to discuss in a paper / Feb 15th

To define what belongs in the results vs. what belongs in the discussion. The results is the report in miniature. This is the data. Examples phrases: …the analysis revealed…does not rule out…the test indicates…seems likely…in contrast…it is highly probable…

Results have:

· Pure numbers—raw data/facts (1,2,5,3,7) and statistical conclusions (mean=4, std error, p-value, etc.). This info goes into the tables (requires column and row headings, title (above the table). Example, “No difference was found between the 2 means (p-0.48, t-test).”

· Graphs illustrate raw data and statistical results (use captions below the figure).

· Text (report comparisons) describes but doesn’t explain data and stats.

Results don’t have:

· Comparisons to other studies—their results aren’t influenced by other researcher’s data in their field

· Context – move any context to the discussion area.

· Background info—mostly the background is inferred from the graphs

A key part in the final paper grade is making sure that information in each section is appropriate (don’t worry about the transitions between sections).

Discussion has:

· Context and no data – it’s all words explaining and interpreting

· Is subjective (as opposed to objective results)

· Conclusion: what your rejection/acceptance of your hypothesis means in the real world. Extrapolate from the specific results of your experiment to a more general question.

· Lots of reference to other scientists work looking for broader implications even if those implications weren’t tested explicitly.

· Building our case for our argument.

· No direct quotes (will deduct 5 points)

· Put in controversies to your results.

· Are you satisfied with the experiment? Admit what you might have been a better approach. What other experiments do you want to do?

Revised text on report:

Summary and conclusion

The testing performed to determine the mechanism of shell thickening indicated that there was no significant difference in shell length growth between the two groups of snails. This is probably the case because the testing was done in close proximity. Other studies have shown that growth patterns in similar levels of genetic variation exist across latitudes (Trussel, 2000)⁴. Surveys of phenotypic variation in 25 Littorina obtusata populations across an approximately 400-km latitudinal gradient in the Gulf of Maine revealed pronounced differences. The shells of snails from northern habitats weighed less and were thinner and weaker in compression than those from southern habitats. In contrast, body size (as measured by soft tissue mass) followed an opposite pattern; northern snails weighed more than southern snails (11–24% for shell mass, 13–17% for shell thickness after 115 days; Trussell). These results are consistent with the expected effects of reduced water temperature on growth (Atkinson, 1994). In addition, increased developmental sensitivity to differences in water temperature may have evolved in northern populations due to the comparably shorter growing season (Conover and Present 1990).

Future experiments that address the role of total body mass induced by food varieties and predator attacks may also help to clarify the mechanisms underlying geographic variation in shell form and body mass. There is a good probability that with studies conducted in locations further apart, and with the use of these variables snail total mass growth will be further exposed.