5-day Introductory PRIMER Training Workshop, 7-11 December 2015
on Multivariate Analysis in Ecology (and other sciences) using PRIMER v7
at Murdoch University, South Street, Murdoch, Western Australia
● A 5-day introductory workshop designed primarily for those who are either new to PRIMER or who have used
only a few standard routines and now wish to understand more of the background to the methods and widen their
experience of potential applications, particularly now that the range of those methods has been expanded with the
launch of PRIMER 7 in 2015. Those who have attended a past PRIMER course and wish only to hear more about
the major enhancements in v7 should instead register for the 3-day WA conversion workshop in the following week.
● PRIMER methods have attracted 20,000 citations in a very wide range of subject areas: marine, freshwater &
terrestrial ecology, oceanography, fisheries, microbial ecology, soil & vegetation science, agriculture, forestry, environmental management & conservation, molecular biology & genetics, toxicology, and many others. The workshop is thus
relevant to any practitioners of multivariate analysis, and does not assume a prior background in statistics – methods
in the PRIMER 7 package are chosen to be non-parametric and intuitively understandable, in a non-mathematical way.
● This training course will cover, at an introductory level, the newer as well as the core techniques in PRIMER 7:
pre-treatment of data (e.g. by transformation or dispersion weighting, and how the choice of these is aided by new
shade plots); appropriate definitions of similarity; agglomerative (and divisive) clustering methods; ordination by
principal components analysis and multidimensional scaling (non-metric nMDS and new metric mMDS options);
permutation testing on similarity matrices, both against prior model structures (the RELATE test) or groupings
(ANOSIM tests, now including an ordered version) and for a priori unstructured data (through SIMPROF tests,
now extended to apply to variables); linking biotic patterns to environmental variables; indices of biodiversity,
including ones based on taxonomic/functional relatedness; means plots (including approximate region estimates,
using bootstrap ideas). A range of plotting features and diagnostic & interpretive tools on MDS plots (animations,
2- or 3-d segmented bubble plots for multiple variables, joint MDS and cluster analysis plots etc) will be met, as will
new wizards, which lead the novice through standard and more complex procedures. This is not a PERMANOVA+
course, but there will be a brief introduction to these add-on hypothesis tests for studies with many designed factors.
● The five days will be a mix of lectures, computer labs and sessions devoted to individual discussions on users’ own
applications. It will be led by Prof K Robert Clarke (Director PRIMER-E, a Plymouth Marine Laboratory fellow, UK
and an adjunct professor at Murdoch University, WA). Bob is a researcher in ecological statistics who has developed
and adapted over several decades the distinctive non-parametric strategy and many specialised methods in PRIMER.
● Participants should bring their own laptops (at least to share in pairs). To allow participants to defer software purchase
until they have tried out the routines on their own data, an authentication key for a downloadable PRIMER 7 (including
PERMANOVA+) will be sent to participants the week before the workshop. This will be only a time-limited licence, and
will cease operation a few weeks after the course (unless of course a purchase has been made, making it time-unlimited).
Note that PRIMER 7 installation is separate from an existing PRIMER 6 and can be run on the same machine without
problems. PRIMER is a Windows-only product, so can only run on a Mac under dual-boot or Windows emulation.
Workshop costs in $AUD (no GST to add)
● Registration fee: $1175 for the 5 days, reduced to $850 for full-time students. This includes a printed methods manual,
the temporary software licence, and mid-am/pm break and lunch catering, but not other meals or accommodation.
● Software purchase: not needed for the workshop, but special $AUD prices are offered to participants only, at:
Private sector: New PRIMER v7 $1000
Upgrade from registered v6 $500
Public sector:
New PRIMER v7 $750
Upgrade from registered v6 $375
University:
New PRIMER v7 $500
Upgrade from registered v6 $250
Also a new full-time student rate: $250 (no upgrade possible since this is a new rate)
Note: a) delivery is by software download, including methods and user manuals in pdf form, on receipt of payment or Purchase
Order – the hard-copy methods manual is supplied at the course and a mono-printed user manual is available at extra cost;
b) the add-on PERMANOVA+ software will only be used for one session in this week, but can be purchased at 50% of the
above prices (if you are already registered with us as a PERMANOVA+ purchaser operating with v6, then there is no charge
to upgrade it to v7 – PERMANOVA+ functionality remains effectively unchanged under v7);
c) if you are making many software purchases (>4 PRIMER) for participants on the same invoice, check with the PRIMER-E
office whether you would get a lower price by taking the GBP bulk purchase rate from the published software price list.
Contact & Registration
● To register for the workshop, please complete the form below and return it to Cathy at the PRIMER-E office, Plymouth
UK (primer@primer-e.com, fax: +44 1752 837721). She can check your current licence status and will issue invoices
for the requested workshop (and software) fees. For queries on local facilities, disability provision, car parking, dietary
needs for the snack lunches etc, contact the local organiser Dr Fiona Valesini, Senior Research Fellow at the Centre for
Fish and Fisheries Research, Murdoch University, f.valesini@murdoch.edu.au, phone: (08) 9239 8805 or 9360 7244.
Or contact the other local organiser, Prof Euan Harvey of Curtin University, euan.harvey@curtin.edu.au
REGISTRATION FORM
5-day Introductory PRIMER v7 workshop
Murdoch University WA, 7-11 December 2015
Please complete a separate form with details for each person wishing to register (though of course you can request a
single invoice covering several registrations and/or software fees).
E-mail/post/FAX the completed form(s) to the PRIMER-E office in Plymouth, UK:
E-mail: primer@primer-e.com
Post: 3 Meadow View, Lutton, Ivybridge PL21 9RH, UK
Fax: +44 1752 837721
Places are limited and will be allocated in order of our receipt of these completed registration details.
Participant details
Name:
Organisational (or home) address:
E-mail:
An invoice will be sent to this e-mail (unless you request a different e-mail address below) for the required fees for
the boxes you cross below [ x ] – see previous page for these costs.
Workshop registration
[ ] Full course fee
or
[ ] Student course fee (full-time student)
Software purchase* (not mandatory for the course – trial versions with one-month expiry will be made available)
[ ] New PRIMER v7 licence
or
[ ] Upgrade PRIMER v6 to v7**
Your current v6 registration (middle 4 characters of key): [
Your PERMANOVA+ registration if any (4-character):*** [
[ ] New PERMANOVA+ add-on licence (PERMANOVA+ only used in one session in this workshop)
]
]
* PRIMER v7 software and pdf manuals are delivered by download. An installation key will be sent on receiving payment for the software or
a Purchase Order covering it. If payment is not received by a week before the workshop – or if software is not ordered – an installation key for
the one-month trial will be made available.
** There is no reduction for upgrading from earlier versions of PRIMER than v6; a new licence needs to be purchased.
*** PERMANOVA+ is not a new version so if we can verify an existing registered PERMANOVA+ licence for you, your installation key for
PRIMER v7 will automatically activate the PERMANOVA+ add-on menu items as well.
Invoice information
[ ] The name/address on the invoice needs to differ from that given above (for the participant).
Please give the required name/address:
[ ] Separate invoices are needed for workshop registration fee and software purchase(s)
[ ] The following Purchase Order number(s) must be given on the invoice(s): [
]
Payment information
The invoice sent for the course fee should be paid by the payment deadline of 16 November 2015 or in 30 days of
the invoice date if that is later. (If this causes problems, please discuss with the PRIMER-E office who are usually
accommodating to requests for fixed extension of payment terms, primer@primer-e.com). Payment can be by bank
transfer, cheque, or credit card, see details on the invoice. An alternative is Paypal, by asking the PRIMER-E office
to e-mail you a Paypal payment request – you do not need a Paypal account to pay in this way, just a credit card.
5-day Introductory PRIMER Training Workshop
Murdoch University WA, 7-11 December 2015
Provisional Programme
Monday, 7 December
09:00-09:15 Introduction
09:15-11:00 Lecture: Measures of resemblance (similarity/dissimilarity/distance) in multivariate structure for
assemblage and environmental data, including shade plots used to assess the effects of pre-treatment
options (e.g. standardisation, transformation, normalisation, dispersion weighting), and guidelines for
different coefficient choices for different data types
11:00-11:15 Coffee break
11:15-13:00 Lecture: Clustering of samples by hierarchical agglomerative (CLUSTER) and brief mention of other
(divisive and non-hierarchical) clustering ideas. Includes discussion of a global test for the presence
of any multivariate structure in a priori unstructured biotic or abiotic samples, using similarity
profiles (Type 1 SIMPROF tests on samples)
13:00-14:00 Lunch break
14:00-16:00 Lab session on transforms, similarity options, clustering and SIMPROF tests*
16:00-16:15 Coffee break
16:15-17:00 Lecture: Ordination (for environmental data) by Principal Components Analysis (PCA)
17:00-18:00 Lab session on ordination by PCA
Tuesday, 8 December
09:00-11:00 Lecture: Ordination (of assemblage data) by non-metric Multi-Dimensional Scaling (nMDS) and
MDS diagnostics (e.g. stress, MST, cluster overlay) for adequacy of low-d representation. Also how
this relates, through the Shepard diagram, to metric MDS (mMDS), useful for abiotic data (Euclidean
distances) and for means plots from biotic data with few means
11:00-11:15 Coffee break
11:15-13:00 Lab session on ordination by nMDS (and mMDS), examining Shepard plots and other diagnostics,
and display tools on MDS plots
13:00-14:00 Lunch break
14:00-15:15 Lecture: Global hypothesis tests of no agreement between two resemblance matrices (RELATE),
comparing assemblage (or environmental) structure with linear or cyclic models in space and time
15:15-17:00 Lab session on RELATE tests for simple seriation without replication and a cyclic model with
replication (coffee break c.16:00-16:15)
17:00-18:00 Lab session on participants’ own data (e.g ensuring it is structured correctly to take into PRIMER*)
Wednesday, 9 December (morning)
09:00-10:30 Lecture: Multivariate testing for differences among a priori specified groups of samples (1-way
ANOSIM and pairwise). Multivariate means plots where global test has established differences. Brief
mention of approximate region estimates for means, in mMDS plots, using bootstrap averaging.
10:30-12:00 Lab session on 1-way ANOSIM and means plots (includes coffee break c.11:00-11:15)
12:00-13:00 Lecture: Ordered ANOSIM tests, 2-way ANOSIM designs and testing (unordered) factors in crossed
ANOSIM designs without replication, by comparing patterns
13:00-14:00 Lunch break
_________________________________________________________________________________________________________
* Throughout, participants will be given real data sets to analyse, but they may also wish to bring their own data. These should be in
numeric, rectangular arrays, with variables (e.g. species) as rows, samples as columns, or vice-versa, in an Excel spreadsheet or
text file. Non-numeric information (factors) on each sample are placed below (or to the side of) this table, separated by a blank
row (or blank column). There is also a 3-column format (sample label, variable label, non-zero entry) suitable for very large
arrays. Opportunity should be taken to discuss your data with the lecturer in labs and breaks prior to the final afternoon session.
Wednesday, 9 December (afternoon)
14:00-15:00 Lab session on ordered and multi-way ANOSIM
15:00-16:00 Lecture: Linking potential environmental drivers to an observed assemblage pattern, via the matching
of multivariate structures (the BEST procedure). Test of no evidence for a biota-environment link,
allowing for selection effects in finding an optimum match (global BEST test)
16:00-16:15 Coffee break
16:15-17:15 Lab session on PCA and BEST for linking to environmental variables, and the global BEST test
17:15-17:40 Lecture: Linkage trees – a further technique for ‘explaining’ assemblage patterns by environmental
variables (LINKTREE), and its relation to unconstrained divisive clustering (UNCTREE)
17:40-18:00 Lab session on link to abiotic variables (LINKTREE) and comparison to unconstrained cluster
Thursday, 10 December
09:00-10:00 Lecture: Species contributions to sample patterns: stepwise form of BEST to identify minimal-sized
species subsets reconstructing the full assemblage pattern (a whole pattern approach), and species
contributions to similarities (SIMPER, a pairwise approach for statistically established groups)
10:00-11:00 Lab session on various methods for identifying species contributions (Matrix display, stepwise BEST,
bubble plots and SIMPER)
11:00-11:15 Coffee break
11:15-12:00 Lecture: Direct analysis of species (or other variables) through species resemblances: techniques for
identifying coherent groups of species (or other variables) in their response across samples
12:00-13:00 Lab session on coherent variable sets
13:00-14:00 Lunch break
14:00-15:00 Lecture: Diversity measures (DIVERSE), multivariate treatment of multiple indices and brief mention
of dominance plots
15:00-16:00 Lab session on DIVERSE, dominance plots, and multivariate analyses of multiple diversity indices
16:00-16:15 Coffee break
16:15-17:15 Lecture: Brief introduction to taxonomic (or phylogenetic) distinctness for simple species lists, as a
biodiversity measures over broad spatial and temporal scales; testing structures (TAXDTEST)
17:15-18:00 Lab session on average taxonomic distinctness with TAXDTEST (if time)
Friday, 11 December
09:00-10:00 Lecture: Brief introduction to Principal Co-ordinates analysis (PCO) and semi-parametric
PERMANOVA test structures (higher-way designs with fixed/random effects in crossed/nested models
with interaction terms) and tests for constant dispersion (PERMDISP), available in the
PERMANOVA+ add-on to PRIMER
10:00-11:00 Optional lab session on PERMANOVA+ methods (or participants’ own data)
11:00-11:15 Coffee break
11:15-12:00 Lecture: Second-stage analysis (2STAGE) to compare taxonomic levels, transformation and
resemblance coefficients
12:00-13:00 Optional lab session on 2STAGE (or participants’ own data)
13:00-14:00 Lunch break
14:00-18:00 Main lab session on analysing participants’ own data using PRIMER (and PERMANOVA+)