diff --git a/src/community/assortativity.jl b/src/community/assortativity.jl
index 0d17ae5f4..4b84f8d1a 100644
--- a/src/community/assortativity.jl
+++ b/src/community/assortativity.jl
@@ -1,21 +1,35 @@
 """
     assortativity(g)
+    assortativity(g, attributes)
 
 Return the [assortativity coefficient](https://en.wikipedia.org/wiki/Assortativity)
 of graph `g`, defined as the Pearson correlation of excess degree between
 the end vertices of all the edges of the graph.
 
+If `attributes` is provided, Pearson correlation is calculated
+from atrribute values associated to each vertex and stored in
+vector `attributes`.
+
 The excess degree is equal to the degree of linked vertices minus one,
 i.e. discounting the edge that links the pair.
 In directed graphs, the paired values are the out-degree of source vertices
 and the in-degree of destination vertices.
 
+# Arguments
+- `attributes` (optional) is a vector that associates to each vertex `i` a scalar value
+`attributes[i]`.
+
 # Examples
 ```jldoctest
 julia> using LightGraphs
 
 julia> assortativity(star_graph(4))
 -1.0
+
+julia> attributes = [-1., -1., 1., 1.]
+
+julia> assortativity(star_graph(4),attributes)
+-0.5
 ```
 """
 function assortativity(g::AbstractGraph{T}) where T
@@ -34,9 +48,25 @@ function assortativity(g::AbstractGraph{T}) where T
     return assortativity_coefficient(g, sjk, sj, sk, sjs, sks, nue)
 end
 
+function assortativity(g::AbstractGraph{T}, attributes::Vector{N}) where {T,N<:Number}
+    P = promote_type(Int64, N) # at least Int64 to reduce risk of overflow
+    nue  = ne(g)
+    sjk = sj = sk = sjs = sks = zero(P)
+    for d in edges(g)
+        j = attributes[src(d)]
+        k = attributes[dst(d)]
+        sjk += j*k
+        sj  += j
+        sk  += k
+        sjs += j^2
+        sks += k^2
+    end
+    return assortativity_coefficient(g, sjk, sj, sk, sjs, sks, nue)
+end
+
 #=
-assortativity coefficient for directed graphs: 
-see equation (21) in M. E. J. Newman: Mixing patterns in networks, Phys. Rev. E 67, 026126 (2003), 
+assortativity coefficient for directed graphs:
+see equation (21) in M. E. J. Newman: Mixing patterns in networks, Phys. Rev. E 67, 026126 (2003),
 http://arxiv.org/abs/cond-mat/0209450
 =#
 @traitfn function assortativity_coefficient(g::::IsDirected, sjk, sj, sk, sjs, sks, nue)
@@ -44,8 +74,8 @@ http://arxiv.org/abs/cond-mat/0209450
 end
 
 #=
-assortativity coefficient for undirected graphs: 
-see equation (4) in M. E. J. Newman: Assortative mixing in networks, Phys. Rev. Lett. 89, 208701 (2002), 
+assortativity coefficient for undirected graphs:
+see equation (4) in M. E. J. Newman: Assortative mixing in networks, Phys. Rev. Lett. 89, 208701 (2002),
 http://arxiv.org/abs/cond-mat/0205405/
 =#
 @traitfn function assortativity_coefficient(g::::(!IsDirected), sjk, sj, sk, sjs, sks, nue)
diff --git a/test/community/assortativity.jl b/test/community/assortativity.jl
index 720668945..9847dabec 100644
--- a/test/community/assortativity.jl
+++ b/test/community/assortativity.jl
@@ -20,4 +20,28 @@ using Random, Statistics
         y = [indegree(g, dst(d))-1 for d in edges(g)]
         @test @inferred assort ≈ cor([x;y], [y;x])
     end
+
+    # Test definition of assortativity as Pearson correlation coefficient
+    # between some scalar attributes characterising each vertex
+    @testset "Small graphs with attributes" for n = 5:2:11
+        g = grid([1, n])
+        attributes = [(-1.)^v for v in 1:nv(g)]
+        @test @inferred assortativity(g, attributes) ≈ -1.
+    end
+    @testset "Directed ($seed) with attributes" for seed in [1, 2, 3], (_n, _ne) in [(14, 18), (10, 22), (7, 16)]
+        g = erdos_renyi(_n, _ne; is_directed=true, seed=seed)
+        attributes = randn(_n)
+        assort = assortativity(g, attributes)
+        x = [attributes[src(d)] for d in edges(g)]
+        y = [attributes[dst(d)]  for d in edges(g)]
+        @test @inferred assort ≈ cor(x, y)
+    end
+    @testset "Undirected ($seed) with attributes" for seed in [1, 2, 3], (_n, _ne) in [(14, 18), (10, 22), (7, 16)]
+        g = erdos_renyi(_n, _ne; is_directed=false, seed=seed)
+        attributes = randn(_n)
+        assort = assortativity(g, attributes)
+        x = [attributes[src(d)] for d in edges(g)]
+        y = [attributes[dst(d)]  for d in edges(g)]
+        @test @inferred assort ≈ cor([x;y], [y;x])
+    end
 end