How to Read a Scientific Paper - aspb.org · Et Scientific governed of conduc must carry integrity,...
-
Upload
nguyendieu -
Category
Documents
-
view
220 -
download
0
Transcript of How to Read a Scientific Paper - aspb.org · Et Scientific governed of conduc must carry integrity,...
How to Read a Scientific Paper
What is a scientific paper? ………………………..…….. 2 Scientific papers are peer‐reviewed....……….....… 3Anatomy of a scientific paper ……………………….... 4How to read a scientific paper ………………………… 5Ethics in research and publishing ………………….... 6Understanding numerical data ……….…………….… 7What does “statistical significance” mean? ....... 8Case study: Buy My Oranges……………..……………..9Appendix: Numerical analysis .....……………….... 12Additional resources and credits ………………… 13
Copyright 2013 American Society of Plant Biologists. www.aspb.org
What is a scientific paper? Scientific papers go straight to the source If someone asks you about a new movie you haven’t seen yet, what do you say? Maybe, “I haven’t seen it, but I’ve heard it’s good.” We generally try to distinguish first-hand from second-hand
information. Like the children’s game of “telephone,” information can change as it is passed along. To get the real story, whether a film review or the results of a new research study, go to the source. Scientific papers present data and interpretations Scientists report the results of their research by writing and publishing scientific papers, which are written in a very formal style. One of the objectives of a scientific paper is to make available the data from a set of studies so that others can learn from them and build on them to address new questions. By publishing and sharing data, scientists work together to advance our understanding. Some articles include results from a few targeted studies, and others present large datasets that other scientists can
use in new ways to address different questions. The authors of scientific papers also provide an interpretation of what they think their new information means and how it contributes to our understanding of how the natural world works. By presenting the data itself as well as the analysis, other authors can evaluate these interpretations for themselves. Because our understanding is always changing, sometimes the interpretations of the data can be re-evaluated in light of new ideas and new data. Society needs scientific literature Much scientific research is publicly funded, and the knowledge and technologies that emerge from research have social impacts. Traditionally, the output from scientific research has been published in journals that are not widely available outside of university libraries, but in the past decade there has been a trend toward increasing openness in science and a desire to make research articles more broadly available. However, it is not sufficient to make these resources available, as in many cases they are written for experts and practicing scientists and therefore not readily comprehensible to those who haven’t been trained in the discipline. We’ve written this article as a guide to help people learn to read the scientific literature, with the goal of increasing access to science and communication about science.
Sc
Peer reviepublicatiois evaluatanonymouused to immay recomcollected well suppo In the stanprocess, sto the righauthors sutheir papejournal edwho evaluwhether this a good for the jouso, the pasent to twmore expewho read provide thfrank evalincluding wsupportedthe study interestingcommentsdecision trevision, o The role oexperimenpresentedthe resultsfind that thrigorous einterpretawhich casrecomme
cienti
ew is a tradn. Prior to ed by otheusly, and t
mprove themmend thaand analyzorted by th
ndard shown ht, the ubmit er to a ditor, uates he topic match
urnal. If aper is
wo or erts, it and
heir luations, whether th
d by the evis novel, im
g. The edits to the auo accept thor to reject
of the reviental designd, as well as. Sometimhe experimenough to stions made
se the reviend that add
ific p
dition in scpublication
er experts, hese evalu
e paper. That additionzed or thathe data be
he conclusividence andmportant, ator forwardthors, alonhe paper, t it.
ewer is to en and the das the intermes, the remental desisupport thee by the auewer mightditional exp
paper
cholarly n, an articlusually
uations arehe revieweal data be t claims noremoved.
ions are d whether and ds the ng with a to request
evaluate thdata rpretation oeviewer wilign was noe uthors, in t periments
rs are
e
e rs
ot
t
he
of l ot
bere M(nnocatoal
ansoexthobdarelathbucu If rethredi
e pee
e performeesults be re
Most of the newspaperot subject tan be effecopics to a blways suffi
nalysis of tometimes xample, th
hat high frubesity, althata to suppesearch repaboratory che rate of gut the medure for can
you want teport on a he original,ead what thiscovered.
er-rev
ed or that tevised.
articles of rs, magazinto peer revctive at intrbroader auciently cau
the results overstate te mass me
uctose cornhough thereport this claport might
conditions, growth of cdia may dencer.
to know mscientific b peer-reviehe researc
view
he analysi
the “massnes, and bview. Althoroducing sdience, the
utious in th
of the studthe conclusedia often n syrup caue is little coaim. Similshow that a new dru
cultured cascribe it as
ore about breakthrougewed articlchers actua
wed
s of the
s media” blogs) are ough they cientific ey are not
heir
dy and sions. For states uses onvincing arly, a under
ug slows ncer cells, s a new
a news gh, find le and ally
Anatomy of a scientific paper
TITLE
AUTHOR INFORMATION
DISCUSSION: An analysis and interpretation of the data presented that integrates the new information with prior findings, states the implications of the work, and sometimes generates new hypotheses tobe tested.
RESULTS: A description of the research conducted and the results obtained.
Results are presented as tables, large datasets, and figures, which can include graphs, videos, diagrams, and photographs.
Some papers include additional supporting data as a supplement.
INTRODUCTION: A statement of what is currently known about the study subject that articulates the questions being investigated. It cites other scholarly works, lays the foundations for the study, and sometimes states a hypothesis to be tested.
REFERENCES: The list of the articles cited in the paper that provide information on the research topic and the methods used.
ABSTRACT: A summary of the study and findings, written by the author.
METHODS: A description of how the studies were conducted, with sufficient detail so that others can repeat them exactly.
Figure legend
Although they are qtime and efew tips toand make Read the The title sresearch, subject or Read the The abstrquestion btaken, andsignificanc Read the The Introdprovide thinformatiogoals of thimportant referenceother relethe text aryear, but scomplete found at threading thhyperlinks Read the Most peoppaper is eunderstanout of ordDiscussiopaper befsection. Tsummarizof the reseexplores t
Howscientific p
quite denseeffort to reao help you e sense of
Title should indic
including trganism.
Abstract ract shouldbeing addrd the majoce.
Introductduction seche necessaon to help yhe study aand interes to previovant work.re often citsometimescitation forhe end of the paper os to the cite
Discussiople find thaeasier to nd if they reer and rea
on section ofore the ReThe Discuszes the findearch and the implica
w to rpapers seee, and it taad one! Heget used tthe paper.
cate the tothe name o
summariz
ressed, theor findings a
tion ction of theary backgroyou undersnd why the
esting. It alous publica The refereted by auths by a numr each refethe paper. nline, thereed referenc
on at the
ead it ad the of the esults sion
dings
ations
read em short, kes a bit oere are a to the form
opic of the of the
ze the e approachand their
e paper wilound stand the e study is so will cite
ations and ences withhor and
mber. The erence is
If you are e are oftences.
a sci
of
at
h
l
e
hin
n
ofaugothco RTexobanDcoprexindemseofso LTadunrehy LA
ientiff the Resuuthors’ cono back to the data on onclusions
Read the Rhe Resultsxperimentsbtained. Bend tables a
Do the dataonclusionsrovides moxperiments
ncluding citescribe sta
methods anections maften find mources or a
ook at thehe cited redditional innderstand eferences cyperlink in
ook at CitAn online ar
fic palts. Once ynclusions athe Resultswhich they
s.
Results ands section ds carried oe sure to eas you read support th
s? The Metore informas and statistations to oandard mend terminolay be unfam
more informa textbook
e Referenceferences cnformation the paper.
can be fouthe online
ting Articlerticle that h
a year lists linhave ciinformabecauspapershow oththe resof the preadingstudiesthese f
aper you’ve readand interprs section toy based th
d Methodsescribes thut and the
examine thd through the authorsthods sectiation aboutstical methother papethods. Somogy in thesmiliar, but y
mation in ot.
ces can lead yothat may h. Often thend through
e article.
es has been por more agks to articlited it. Thisation is usese from the you can fiher authorults and co
paper you g and how s have extefindings.
d the retations, o examine eir
s he data e figures the text. ’ ion t the
hods, rs that me of the se you can her online
ou to help you ese h a
published go usually es that s eful ese ind out rs interpret onclusionsare other
ended
s
Et
Scientific governed of conducmust carryintegrity, fmiscondufalsificatiomake hea Journal eda code of suspend tobjectivelyreviewer hwork as calong the reviewer awork in th“scoop” owork.
Scientific public’s trresearch, wasted efunsubstanclaims. Mdevelopedethical gufor trainingthese are
thics
research aby ethical
ct. Researcy out their fairness, act, such as
on, or plagiadlines.
ditors and conduct th
their persoy as possibhas to treaonfidentialinformatio
also has toe paper, er contradic
misconducrust, disrupand if und
fforts as otntiated, falsany organd materialsidelines ang young sclisted belo
s in r
and publishprinciples
chers and awork with nd respects data fabrarism, is ra
reviewers hat requireonal biasesble. For ex
at the unpul and cann
on to otherso fairly evaeven if it thrct the revie
ct undermipts the progdetected, chers try to se, or fabrizations has that explind that cancientists; aow.
resea
hing are and a cod
authors honesty, t. Serious rication, are, but ca
are held tos they
s and act axample, a blished ot pass s. The luate the reatens to wer’s own
nes the gress of an lead to build on icated ave citly define
n be used few of
arch a
de
an
o
s
e
RTh
Th
Th
Th
Th
and p
Resourceshe America
has develoin publishi(http://wwwthics.cfm)
he US Offichttp://ori.htraining tointeractivein which thmakes chomiscondu(http://ori.h
he Commit(COPE; htprovides gconduct foreviewed j
he Scientifichas a wearesearch a(http://wwwdex.shtmlpublished is availabl(http://wwworTheRec
he U.S. Nathas publisScientist: Ain Researincludes indiscussion(http://www=12192).
publ
an Society oped a suiting: w.aspb.org/. ce of Reseahhs.gov/). Oools the ORe video dramhe viewer soices as anct unfolds hhs.gov/Thttee on Pubttp://publicaguidelines, tor editors anjournals. c Researchalth of resouand publishw.sigmaxi.o). A set of sin its journa
le as a free w.sigmaxi.o
cord.pdf). tional Acadshed a bookA Guide to
rch, which isnteresting cn w.nap.edu/
ishin
of Plant Be of policies
/publication
arch IntegrOne of the mI has develoma called “Telects a cha
n issue of re
eLab/TheLablication Etationethics.otraining, annd publishe
Society Sigurces abouthing org/programsix essays oal AmericanPDF
org/program
demies of Sk called On Responsibl
s free to dowcase studies
catalog.php
ng
iologists s for ethics
ns/editoriale
rity (ORI; many oped is an The Lab,” aracter and
esearch
ab.shtml). thics org/) d codes of
ers of peer-
gma Xi t ethics in
ms/ethics/inon ethics n Scientist
ms/ethics/F
Sciences Being a le Conduct wnload ands for
p?record_id
e
d
d
d
Scientistsmethods iinclude nuscientist wtolerant soof seeds pdrought-tothe amouthose see Numericabe easily misrepresanalyze nappropriathe data, tvalues, anWhen meindependeused to dealso how sample po The arithThe arithmusually haan averagthe value dividing thmeasurem Sometimecoinciden(the moderange of valways. Fthe three (indicated
Unds use manyin their inveumerical dawho wants oybeans mproduced bolerant varnt of oil or
eds.
al data are misunders
sented. Theumerical dte for the mthe distribund the para
easuremenent sampleetermine thmuch variaopulation.
metic meametic meanave in mindge. It is calof each m
he sum by ments.
es the meatly be the m
e) or fall rigvalues (theor exampledatasets s
d by a red a
erstay complemestigationsata. For exto develop
might countby the stanriety, or migprotein co
very powestood or e methods
data must bmethod useution of theameters mts are colle
es, the valuhe averageation there
an n is the vald when weculated byeasuremethe numbe
an value camost commght in the me median), e, the mea
shown beloarrow), but
andinentary s and oftenxample, a p drought-t the numbndard and ght quantifntained in
erful but ca
s used to be ed to collee data easured. ected fromues can bee value ande is in the
lue we e talk abouty adding upnt and er of
an also mon value middle of thbut not
an of each ow is 10 t the range
ng nu
n
ber
fy
n
ct
e d
t p
he
of
e
andifrex-amemaco ThThsimmethearstada13rigthestarigtheasanbesta
Thcaa sAp
Cldoinfvadeinc
umernd distributfferent in epresented axis showseasuremenany times
ounted).
he standarhe standarmply calledeasuremene set of daround the mandard devata sets be3, and 23, fght. The moe data, theandard devght, we’ve e mean of s a bar of hn error bar eyond the mandard dev
he simplesalculate thespreadsheppendix for
early, a sinoes not conformative o
ariation of teviation, is cluded in a
rical tion of the each samp
as a histos the valuent, and thethat measu
rd deviatiod deviationd the standnt of how m
ata varies mean. The viations of
elow are 3.2from left toore spread
e larger theviation. Onrepresenteeach sam
height 10, wthat extenmean by oviation.
t way to e standard eet programr more info
ngle value nvey enougon its own.the data, salso nece
a research
5 10 15 20
datavalues arele. (These
ogram, in we of the e y-axis shourement w
on of the mn of the medard deviatmuch
the 2,
o d out e n the ed ple with ds
one
deviation m like Exceormation).
such as thgh informa. A measuruch as thessary and article.
25 30 35 40 45
e very
data are which the
ows how was
mean ean (often tion) is a
is to use el (see the
he mean ation to be re of the e standard
should be
5 50 55 60 65 700 75
Wha SignificanWhen comthe differesmall but ndeterminemeaningfua measurestatistical tdifference “statisticadoes not mjust that thsamples a Taller, or As an exaon the eason the wesextremely every persmeasure aa statistica We measufood restatown and cdeviation odeviation ifollowing ttown, the m163.0 cm (11). TheyTo find ouis commonrelated buStudent’sExcel; see By convenprobabilityThe null hysamples ahypothesdifferent.
at doe
nt, but statmparing samnces in menevertheles if observe
ul, scientisted differenctest, it can between th
ally signifimean that there is a strare different
not? mple, let’s st side of tost side of todifficult to
son in the ta subset of al test to he
ure the heigurants on tcalculate thof each samis often indthe mean). mean of m(11) and on
y don’t lookt, we can unly used tot unlinked d
s t-test (whe the Appen
ntion, a t-tey that the nypothesis s
are the samsis is that thThe output
es “st
tistically smples or poeasured valss meaningd differences use statisce is suppobe said thahe samplesicant.” Notthe differenrong probat.
see if peopown are tallown. It woumeasure thown, so insf people anelp us interp
ght of 20 pethe east or he mean anmple (the sicated in paOn the weeasured hen the east,
k different, buse a statis compare tdata, called
hich can be ndix).
st calculateull hypothstates that
me. The altehe two samt of a t-test
tatisti
significant?opulations, lues can begful. To es are stical tests.orted by theat the s is te that this nce is real, bility that th
ple who liveler than thould be he height ostead we wd then apppret the da
eople at faswest side o
nd standardstandard arenthesesst side of eights is 163.3 cm
but are theytical test thtwo sets of d a done in
es the hesis is true
the two ernative
mples are is a p-valu
ical si
?
e
If e
he
e ose
of will
ly ta.
st-of d
s
y? hat
e.
ue,
whhytypThbehyrej WsaThhynotheresres Win (Wstaat cotheoftesThsapestagravaforshmesahig StJumenesh
ignifichich is the pypothesis ispically rejechat is, unlesetween the ypothesis (tjected.
When we apamples, theherefore, wypothesis, ao statisticalle heights ostaurant (Wstaurant (E
We can also a school ca
WS). In thisandard devthe school
ompare these west-side163.0 cm (st to these herefore, way that the deople at WRatistically saph to the
alues and sr the three
hown. To ineasured va
ample differghlight them
tatistically ust becauseeaningful dewsworthy thorter than
canceprobability s true. By cct the null hss p<0.05,samples, a
that the sam
ply the t-tese calculatede do not reand we conly significan
of people atWR) and theER).
measure tafeteria in t
s sample, thviation of th are 145.4 se values te restauran(11). Now wsamples, te reject thedifference iR and at Wignificant. Iright, the mtandard desamples adicate that
alues in ther significantm with a sta
y significane a result isdoesn’t makto find that adults?
e” methat the nuonvention, hypothesis there is no
and the altemples are d
st on our hd p-value iseject the nunclude that nt differenct the west-se east-side
he height othe west sihe mean anhe heights m
cm (18). Wto those ment, which hawhen we ahe result is
e null hypotn height be
WS is In the
mean eviations re
e WS tly, we ar.
nt, but signs statisticallke it importschool kid
an? ull
biologists if p<0.05.
o differenceernative different) is
eight s 0.5. ll there is
ce betweenside e
of people de of town nd measured
We can easured at ad a mean pply the t-s p<0.05. thesis and etween
nificant? ly ant. Is it s are
*
e
s
*
You like frlike to getI have devoranges, based on C content
You’ve alworanges, market, yovariety. Tooranges a
A and a Bboth. Thedon’t noticThey alsovolume of
Now I useand an indvitamin C amount ofyou the re18.3 mg oand B hasml juice.
Given thagood, andamount ofvitamin Cof A? Why
Casresh orangt some vitaveloped a BETTER (their highe
t.
ways bougbut when you notice mo show youare, I squee
B orange any’re both dce any diffe
o produce ef juice per o
e my portabdicator dyeto give a cf vitamin C
esults. Theof vitamin Cs 19.8 mg o
at the juice d the orangf juice, but, will you sy or why no
se stuge juice, anamin C in thnew varietB), that I’mer-than-usu
ght AVERAyou go to tme selling u how amaeze the jui
nd let you delicious, aerence in texactly theorange.
ble spectroe, which recolor that in
C in the juice data showC per 100 mof vitamin
from B tasges product B producestart buyingot?
udy: nd you alsohe morningty of
m marketinual vitamin
AGE (A) he farmersthe new
azing my ce from an
taste themand you their flavore same
ophotometeacts with ndicates thce. I show w that A haml juice, C per 100
stes just asce the sames more g B instead
Buyo g.
g n
s
n
m
r.
er
he
as
s me
d
Ifmyyy
my of I haven
my orangyou? Whayou haveyour ques
oran’t convin
ges, how at other q? How wstions an
ges nced you
can I coquestion
would younswered
u to buy onvince ns do u get
d?
Are they You probadifferencevitamin C (B has abwonder if random. Mpicked a Bthan-averprobably wobserved small differeal mean
It is possiquestionscontent frovariety anstandard d
Let’s look two types different athe null hyno differehypothesidifferent. content of
really diffably are coe between
in the twobout 8% mo
that differeMaybe I jusB orange wrage level owant to knois repeata
erence of lening.
ble to answs by measuom severa
nd analyzindeviation o
at two sce of orange
and I just gypothesis, nce. In theis, the two If we measf lots of diff
ferent? oncerned ththe amoun juices is faore than A)ence is reast got luckywith a slighof vitamin Cow if the d
able and whess than 1
wer both thuring the vial oranges ng the meaof both dat
enarios. In s are not rot lucky. Wmeaning t
e alternativtypes of o
sure the vitferent oran
hat the nts of airly small ) and al or y and
htly higher-C. You ifference I hether a 0% has an
hese tamin C of each
an and asets.
one, the really We’ll call ththat there ive ranges aretamin C nges, we
ny
hat s
e
casc
Avityscthvavaacin
Instofanth1819Scoserep<fop=
an begin tocenarios is
A histogramtamin C va
ype B (orancenarios ishe data, wealues for Balues for Actually ass
nterpretatio
n the first dtandard def A and B rnd 19.09 (0he values a8.39 (0.39)9.44 (0.35)tudent’s t-tonfirms thaecond dataeally is diffe<0.05, whe
or the first s=0.5.
o figure ous supported
m plot showalues from nge) for thes shown bee can get aB tend to beA in the botsign a certaon by analy
dataset, theeviation shorespectivel0.67). In thare ) and ). The test at the aset erent: ereas set,
1
t which of d.
wing the me type A (ble two differelow. Just la sense thae higher thttom plot, bainty to thisyzing the n
e mean (wiown in parly are 18.9he second
8 -
these
easured ue) and rent looking at at the han the but we cans
numbers.
ith the rentheses) 95 (0.67) dataset,
*
Therefore, by repeating the measurement, we can work out whether the two samples are different or the same. Because scientists regularly measure small but meaningful differences, replication, numerical analyses, and statistical testing are essential tools for scientists. Other considerations Which of these statements are supported by this study? B oranges are healthier. If I eat B oranges I will get fewer colds. B oranges are a better value. B oranges always have more
vitamin C. The answer is that none of these statements are supported by the data presented in this study. Based on the distribution of values shown in the second, “alternative hypothesis,” dataset our study showed that there is slightly more vitamin C on average in the B oranges, but any claims beyond that are unjustified extrapolations. To investigate these issues, we would have to design and perform additional experiments. We could also examine the scientific literature to investigate the relationship between vitamin C consumption and the incidence of colds. Furthermore, the characteristics of a fruit are somewhat variable and
depend on the conditions in which it was grown. Perhaps next year the vitamin C level will not be different between A and B. It would be nice to see a difference that extends over more than one harvest season and more than one growing region. Science provides us with a set of very powerful tools with which to understand the world, but our knowledge accrues in small steps. Any single study provides an incremental advancement, but also lays the groundwork for further studies. Caveat emptor If you wanted to buy a car, you wouldn’t take the advice of the salesperson on the car lot, would you? You would first do a side-by-side analysis of the cars you like to compare fuel efficiency, safety ratings, and the reputation of the manufacturer. An educated consumer makes good choices by looking at evidence. We are all consumers of science, whether it informs our decisions about voting, health care, eating, purchases, energy use, or merely how we understand the world. Although the scientific literature can be a bit daunting, it is an important resource. Learning to read scientific papers empowers you to cut through the hyperbole and hype, to see for yourself the evidence that lies behind the claims, and to draw your own informed conclusions from the studies.
Appendix – Numerical analysis Here are the numbers graphed on the page titled “Understanding numerical data.”
Graph 1 Graph 2 Graph 3
4 0 0
5 0 0
6 0 0
7 1 0
7 1 0
8 1 0
8 2 0
8 2 0
9 2 0
9 2 0
9 4 0
10 6 0
10 6 0
10 7 0
11 9 0
12 10 0
12 12 0
13 16 0
13 20 0
14 20 37
14 22 51
15 33 66
16 54 76
Mean 10 10 10
StDev 3.23 13.01 23.16
The mean and standard deviation were calculated using functions in Excel (2010). To calculate the mean of a group of numbers, in an empty cell below them, type “=average(…)”. The space between the parentheses is filled with the first and last cell in the column—for example, (A6:A29). After you type the left parenthesis, you can use your mouse to draw a box around the column of numbers and then type the right parenthesis. To calculate the standard deviation, type into an empty cell “=stdev(A6:A29).”
Here are the numbers used on the page titled “What does ‘statistical significance’ mean?”
East Rest. West Rest.
West
School
154 182 125
166 166 140
172 159 136
148 155 160
158 143 125
173 177 150
161 156 158
180 172 136
175 168 124
155 165 180
159 145 172
183 156 124
174 154 118
165 157 156
147 172 145
164 167 136
167 154 145
177 177 154
150 184 167
173 158 158
Mean 165.05 163.3 145.4
StDev 10.80 11.54 17.63
t‐test 0.633 (ER vs. WR)
0.005 (WS vs. WR)
The mean and standard deviation were calculated as described previously. The t-test was also calculated using a function in Excel. Click on the function box (fx) icon at the top of the page. In the pop-up window, select the category “statistics” and then T.Test. In the pop-up box, select one set of numbers for Array 1, the second for Array 2. For this kind of data, (independent, Gaussian distributed), select “2” for tails and “2” for “type.”
Additional Resources and Credits Additional Resources American Statistical Association: Education.
(http://www.amstat.org/education/onlineresources.cfm)
Benchpress Project. (2012). Resources to teach journalists about scientific papers and statistics (http://www.benchpressproject.org.uk/)
Hans Rosling. (2010). The Joy of Stats (Video) (http://www.gapminder.org/videos/the-joy-of-stats/)
Sense About Science Making sense of statistics (http://www.senseaboutscience.org/resources.php/1/making-sense-of-statistics)
Peer review: The nuts and bolts (http://www.senseaboutscience.org/resources.php/99/peer-review-the-nuts-and-bolts)
I don’t know what to believe: Making sense of science stories (http://www.senseaboutscience.org/resources.php/16/i-dont-know-what-to-believe)
University of California Museum of Paleontology Understanding Evolution The journal club toolkit
http://evolution.berkeley.edu/evolibrary/teach/journal/index.php
Understanding science: How science really works The social side of science: A human and community endeavor
(http://undsci.berkeley.edu/article/0_0_0/socialsideofscience_01)
Publish or perish? (http://undsci.berkeley.edu/article/0_0_0/howscienceworks_15)
Scrutinizing science: Peer review (http://undsci.berkeley.edu/article/howscienceworks_16)
A blueprint for scientific investigations (http://undsci.berkeley.edu/article/0_0_0/howscienceworks_03)
Visionlearning Scientific communication: Understanding scientific journals and articles (http://www.visionlearning.com/library/module_viewer.php?c3=&mid=158&l=)
Scientific communication: Peer review (http://www.visionlearning.com/library/module_viewer.php?mid=159)
Scientific communication: Utilizing the scientific literature (http://www.visionlearning.com/library/module_viewer.php?mid=173&l=)
Scientific research: The case of the ivory-billed woodpecker (http://www.visionlearning.com/library/module_viewer.php?mid=174&l=)
Produced by Mary Williams for the American Society of Plant Biologists (www.aspb.org). About the American Society of Plant Biologists (ASPB) WHAT IS ASPB? ASPB is a professional society devoted to the advancement of the plant sciences. It publishes two world-class journals and organizes conferences and other activities that are key to the advancement of the science. Membership in the American Society of Plant Biologists is open to anyone from any nation who is concerned with the physiology, molecular biology, environmental biology, cell biology, and biophysics of plants and other related matters. WHAT IS A PLANT SCIENTIST? A plant scientist specializes in the scientific study of plants. Within plant biology there are many areas of interest, including cellular and molecular biology, genetics, development, evolution, physiology, and biochemistry. Plant scientists are working worldwide in nearly all sectors including academia, corporations, pharmacology, research, nonprofits, and government.