PRIMER Reference

Q: What is PRIMER software used for?

PRIMER (Plymouth Routines In Multivariate Ecological Research) is specialized software for multivariate analysis of ecological community data. It is the standard tool for analyzing species abundance matrices from marine, freshwater, and terrestrial surveys. Key analyses include nMDS ordination, ANOSIM/PERMANOVA hypothesis testing, SIMPER species contributions, and BEST/BIO-ENV environmental matching.

Q: What is the difference between ANOSIM and PERMANOVA?

ANOSIM is a non-parametric rank-based test using the R statistic (-1 to +1) to assess group differences. PERMANOVA is a permutational ANOVA that partitions multivariate variation using Pseudo-F statistics and supports complex experimental designs with fixed, random, and nested factors. PERMANOVA is more powerful for multi-factor designs, while ANOSIM is simpler for one-way comparisons.

Q: How do I interpret nMDS stress values?

Stress below 0.05 indicates an excellent representation of the data in 2D. Stress below 0.10 is good, below 0.20 is still useful and interpretable, but above 0.30 the ordination is essentially random and should not be interpreted. If stress is high, try increasing dimensionality to 3D or reducing the number of samples.

Q: When should I use Bray-Curtis vs Euclidean distance?

Bray-Curtis similarity is the standard for biological abundance data because it ignores joint absences — two sites are not considered similar just because they both lack a species. Euclidean distance is appropriate for continuous environmental data (temperature, salinity, depth) which should be normalized before calculating the distance matrix.

Q: What does SIMPER analysis tell you?

SIMPER (Similarity Percentage) identifies which species contribute most to within-group similarity and between-group dissimilarity. It reports each species' average contribution percentage and cumulative percentage. A species with Avg. Diss / SD ratio greater than 1.4 is considered a good discriminating species between groups. Typically the top 5-10 species explain the majority of the pattern.

Q: How do BEST/BIO-ENV and DistLM differ?

BEST/BIO-ENV finds the subset of environmental variables whose distance matrix best correlates (Spearman rho) with the biotic similarity matrix. DistLM (Distance-based Linear Model) quantifies how much variance in the community data each environmental variable explains, using model selection criteria like AICc or adjusted R-squared. DistLM results can be visualized with dbRDA constrained ordination.

Q: What data transformation should I use?

The choice depends on your analysis goal. No transformation emphasizes dominant species. Square root transformation is the most common general-purpose choice. Fourth root down-weights dominants further, giving more influence to rare species. Log(X+1) is useful for highly skewed data. Presence/absence transformation treats all species equally regardless of abundance.

Q: Is this PRIMER reference free?

Yes, this PRIMER reference is completely free with no usage limits, no account registration, and no software installation. All content is rendered in your browser and no data is uploaded to any server. It is designed as a practical quick-reference companion to PRIMER v7 and PERMANOVA+.

Free reference guide: PRIMER Reference

25 results

About PRIMER Reference

The PRIMER Reference is a detailed, searchable guide to PRIMER v7 (Plymouth Routines In Multivariate Ecological Research) and its PERMANOVA+ add-on. It covers the full multivariate analysis workflow from data preprocessing and transformation through similarity calculation, ordination, cluster analysis, statistical hypothesis testing, species contribution analysis, and environmental association modeling.

Built for marine biologists, community ecologists, environmental scientists, and graduate students, this reference provides quick access to Bray-Curtis similarity, nMDS ordination with stress interpretation, ANOSIM and PERMANOVA hypothesis tests, SIMPER species contribution analysis, BEST/BIO-ENV environmental matching, DistLM and dbRDA constrained ordination, and diversity indices including Shannon, Simpson, and taxonomic distinctness.

Every entry includes the exact PRIMER menu path (e.g., Analyse > MDS > non-metric MDS), parameter interpretation guidelines, and practical decision criteria. All content is organized into categories: Overview, Preprocessing, Similarity, Ordination, Cluster Analysis, Statistical Tests, Species Contribution, Environmental Association, Visualisation, and Diversity.

Key Features

Complete PRIMER v7 menu-path reference for every analysis — from Analyse > MDS to PERMANOVA+ > DistLM
Bray-Curtis and Euclidean distance formulas with guidance on when to use each similarity measure
nMDS stress value interpretation thresholds (excellent < 0.05, good < 0.10, useful < 0.20)
ANOSIM R statistic and PERMANOVA Pseudo-F interpretation with pairwise post-hoc test guidance
SIMPER species contribution breakdown — average dissimilarity, cumulative %, and discriminator ratios
BEST/BIO-ENV and DistLM for linking environmental variables to community patterns with variable selection criteria
Diversity index calculations including Shannon H', Simpson 1-Lambda', Margalef d, and Pielou J'
Shade plot and bubble plot visualization techniques for species-sample matrix pattern detection

Frequently Asked Questions

What is PRIMER software used for?

PRIMER (Plymouth Routines In Multivariate Ecological Research) is specialized software for multivariate analysis of ecological community data. It is the standard tool for analyzing species abundance matrices from marine, freshwater, and terrestrial surveys. Key analyses include nMDS ordination, ANOSIM/PERMANOVA hypothesis testing, SIMPER species contributions, and BEST/BIO-ENV environmental matching.

What is the difference between ANOSIM and PERMANOVA?

ANOSIM is a non-parametric rank-based test using the R statistic (-1 to +1) to assess group differences. PERMANOVA is a permutational ANOVA that partitions multivariate variation using Pseudo-F statistics and supports complex experimental designs with fixed, random, and nested factors. PERMANOVA is more powerful for multi-factor designs, while ANOSIM is simpler for one-way comparisons.

How do I interpret nMDS stress values?

Stress below 0.05 indicates an excellent representation of the data in 2D. Stress below 0.10 is good, below 0.20 is still useful and interpretable, but above 0.30 the ordination is essentially random and should not be interpreted. If stress is high, try increasing dimensionality to 3D or reducing the number of samples.

When should I use Bray-Curtis vs Euclidean distance?

Bray-Curtis similarity is the standard for biological abundance data because it ignores joint absences — two sites are not considered similar just because they both lack a species. Euclidean distance is appropriate for continuous environmental data (temperature, salinity, depth) which should be normalized before calculating the distance matrix.

What does SIMPER analysis tell you?

SIMPER (Similarity Percentage) identifies which species contribute most to within-group similarity and between-group dissimilarity. It reports each species' average contribution percentage and cumulative percentage. A species with Avg. Diss / SD ratio greater than 1.4 is considered a good discriminating species between groups. Typically the top 5-10 species explain the majority of the pattern.

How do BEST/BIO-ENV and DistLM differ?

BEST/BIO-ENV finds the subset of environmental variables whose distance matrix best correlates (Spearman rho) with the biotic similarity matrix. DistLM (Distance-based Linear Model) quantifies how much variance in the community data each environmental variable explains, using model selection criteria like AICc or adjusted R-squared. DistLM results can be visualized with dbRDA constrained ordination.

What data transformation should I use?

The choice depends on your analysis goal. No transformation emphasizes dominant species. Square root transformation is the most common general-purpose choice. Fourth root down-weights dominants further, giving more influence to rare species. Log(X+1) is useful for highly skewed data. Presence/absence transformation treats all species equally regardless of abundance.

Is this PRIMER reference free?

Yes, this PRIMER reference is completely free with no usage limits, no account registration, and no software installation. All content is rendered in your browser and no data is uploaded to any server. It is designed as a practical quick-reference companion to PRIMER v7 and PERMANOVA+.