MbShea@PWP@English@UMCP

WEDNESDAY CLASS IS CANCELLED.  I am watching for COVID and will test in a few days.  I had a very close contact on Sunday and now that family is all positive. 

We are now focusing on the ending of the one-article review (three or four paragraphs as you slide off the lemon shaped knowledge or explore briefly the heavy end of the pear-shaped knowledge. One of those paragraphs concerns the logos of numbers as

1. an argument that clarifies the conclusion to be drawn from the data analysis of findings OR (and)
2. additional evidence to vett the argument being made OR (and)
3. a check on the hypothesis and claim thinking (remember that we test the null hypothesis in many cases) that means we do not think the finding is due purely to chance or randomness.

Here is some background knowledge on the stats paragraph.  Generally, I want you to understand that the power of stats and numerical analysis is not without limits. 

 I urge you to talk about statistics with your science professors.  For example, in my field of ecology and environmental science, we are in a quiet riot over frequentist, mutivariate, and Bayesian statistics.  This is an assigned reading for me, in one of my classes. Here is another.

For biomedical researchers, you may appreciate this analysis of the limits of p-values in biomedial research.

Please look at your research articles for Friday's Eli Review post, and note the type of statistics tool used.  Look this up in some way to have a working definition.  Common tools or tests include:

• p-values
• Confidence intervals
• Student's t test (and corrections)
• Analysis of variance (ANOVA)

I simply want you to know about this area within science articles, even if you do not understand the statistics. You would not be alone among scientists, if you don't.  I don't.  However, I want you to leave this class with an understanding of this quandry.  And, the limited definition of significance testing and p-values.  For fun, enjoy this comic.

I would write the stats para in first person.  Here are some sentence starters:

• I agree with the use of two-tailed ANOVA testing in this study.....
• I appreciate the author's carefulness when dealing with small sample sizes iin categorical variables, hence the use of the Egon-Pearson correction.
• I notice that the linear regression visual is presented twice: once with all sample values included and accompanied by one with seven NOx readings excluded, due to instrument error (incorrect tare settings on that date).

Note: this type of writing and analysis are often NOT part of formal presentations, say, in a conference.  However, this type of discussion IS PART OF GRADUATE SCHOOL/PROFESSIONAL SCHOOL course work in seminars.  Journal club culture also tends to include this type of discussion orally because this is how we learn.  Always be learning (ABL).  This is hard wired into the life of science/technology. This is why we practice such thinking and working. 

Now, onto other paragraphs at the end of your review when you talk about meaning! Let's gather some useful boilerplate language from Man U Academic Phrasebank:

Expressing a causal relationship tentatively

Finally, read this excellent blog post by water policy expert./professor Raul Pacheco Vega. He summarizes -- including with visuals -- an excellent book about excrement: Josh Bernoff’s Writing Without Bullshit: Boost Your Career by Saying What You Mean. 

Friday, good morning.  I will be in as per usual at Google Meet link: 9:9:50, 10-10:50, 11-11:50.  Here is some language helpers as you turn away from the three point paragraph to your closing section:

(if you analyzed WITHIN the body paragraphs you can say) I assessed some aspects of Salazar's findings earlier.  However, I want to note some of the limitions of this work offered by the team.  One concern expressed about this observational study is....

Let's look briefly at the use of a one-tailed ANOVA test in this study. One-tailed tests are much less common than the two-tailed test of variance.  I think that one reason for this use is that the underlying distribution in these treatement groups are very different, especially in the left hand tail.

The researchers report their p-values at both the 90% and 95% levels, which is helpful to researchers who may want to continue this inquiry albeit with a modified study design.

Now, let's talk about the other ways to close up.  You can return to the opening strategies for some content or ideas.  You can also focus on:

• suggested reading
  • other studies that this work references
  • Science Daly articles that summarize related research
• immediate applications for diagnosis or treatment (many of you are looking at biomedical research)
• how basic knowledge is built in fields like physics, math, astronomy, and, say, philosophy
• how the study helps with an ongoing, large social problem that needs knowledge, modeling, or proposed/tested solutions.

(Thursday) Good morning, I just found this, which is an inverted take on my lemon/pear approach to how science research articles are shaped. From this PDF guide (101 pages) developed by the Asian Institute of Technology.