A systematic review was conducted to assess the use of statistics in the field.
I think you mean in this field.
It was determined that the field and influence of restoration ecology had increased dramatically since its development.
Does this mean the research?
There were a few less common tests used in some of the studies.
Be specific and state the proportions directly.
Better description of how the stats were actually done is a superb idea.
First paragraph sets the stage well. Could use a bit of editing for style and flow.
The second paragraph should be not be stand-alone. Shorten-first para and work second one into it as the need change in the field.
Too many citations to Michener, can you find other more recent sources to back up your statements.
change in this study to in this review, formal systematic review techniques were used to examine... etc.
Reword these sentences: The primary purpose of this paper is to assess how statistics are used in restorations studies. This was determined by examining the following objectives:
you have questions not objectives. Purpose and objectives are often interchangeable to either change to Qs or find different terms.
Cut predictions afterwards. Redundant.
cut: To get an overview of the use of statistics in restoration ecology, the literature of the field was systematically reviewed.
Why did not also not check the term arid?
never begin a sentence with a number: 21 different restoration factors were examined... change to A total of... etc.
Change fig to Fig.
Do not state the state but the finding instead and put the statistical test in parentheses. State the findings and the stats support as the evidence.
A Chi-squared test was performed and found that the distribution of tests was significantly different than expected (p<0.0005). ANOVA tests were used much more often than any other test (n=71).
Change to The frequency of statistical tests used in restoration ecology in deserts was significantly different with ANOVA as the the most common (Chi-square test, test statistic = ?. p = 0.0005, with post hoc test for ANOVA).
revise this bit: Citation rate also varied with test. The 6 most common tests were cited the most. confusing
Cut: Impact factor was compared by year. this is a method.
While the average impact factor has remained the same over the years, the range of impact factors has become wider more recently. ? Cite figure or stats....
Be specific with all findings.
Figures are great. Cut labels on top of plots. Figure legends also need more detail.
I like this paragraph:
Statistical test of some kind or another were used in a majority of the studies (fig. 6). This is despite the fairly frequent use of observational studies in restoration (Zhang et al. 2012). This is likely due to the usefulness of statistics when assessing change. Without some kind of comparison it is harder to quantify change and determine how effective the restoration technique at hand is. Statistical analysis also allows for more comparisons to be made over more sites or repeats such as Peters et al. with 4800 repeats (2008). Studies with statistics tended to have larger sample sizes than those without (mean sample size 4.4 repeats for sites without statistics versus 489.3 repeats for studies with) (Noble 2014). It is not feasible to compare more than a few sites by observation alone. Statistics allow a wider area to be surveyed and increase the scope of the study.
However, this is more recent and direct evidence you can cite here. Check the latest issue of Frontiers in Ecology & the Environment.
Do not re-cite figures in the Discussion. Discussion should be interpretation not a repeat of results section please.
Conclusions are good.
Just give it a nice edit (discussion) and streamline the writing a bit and I think you are in good shape.
ALSO, make this paragraph specific to your field.
As stated by Michener et al., restoration studies often have a less than typical experimental design, which can drive the need for different statistical tests (1997). Most studies can be analyzed by ANOVAs or linear regressions, the two major types of statistics found in ecology (Cairns et al. 1996, Michener et al. 1997). Arid area restoration studies seemed to be similar. The most common statistical tests of the studies support this, with ANOVA, t-tests, and linear regressions making up a large portion of the total number of tests (fig. 1). In addition to being appropriate for common study designs, many ecologists are also familiar with these tests, and may feel most comfortable using them. Researchers may never have studied nonmetric multidimensional scaling, but most science undergraduate studentsget a basic training to these common tests in an introductory statistics class by (Michener et al. 1997). The use of common statistics can be beneficial, as a wide range of individuals will be able to interpret the results easily. This is provided of course that the statistics chosen are appropriate for the study design. There are example where more powerful or more appropriate tests should be used (Michener et al. 1997). The results of this study show that there are many other kinds of tests that can be used in restoration ecology. To bring ecology terms to the statistics there is good diversity to the tests but not much evenness. The more widespread use of some of these “uncommon” tests might benefit the field more than the dominance of a few tests if they are used correctly.
It is likely really important for managers and applied scientists using this lit to have the stats better handled and the most appropriate tests used. Was there much variation in the designs? Is the diversity of the stats an indication of the experimentation underlying this field?