The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
trajectories ships with Nestimate: 138 student sequences
× 15 weeks, three states (Active, Average,
Disengaged) plus NA for missed weeks.
data(trajectories)
dim(trajectories)
#> [1] 138 15
head(trajectories[, 1:8])
#> 1 2 3 4 5 6
#> [1,] "Active" "Disengaged" "Disengaged" "Disengaged" "Active" "Active"
#> [2,] "Average" "Average" "Average" "Average" "Average" "Average"
#> [3,] "Average" "Active" "Active" "Active" "Active" "Active"
#> [4,] "Active" "Active" "Active" "Active" "Active" "Average"
#> [5,] "Active" "Active" "Active" "Average" "Active" "Average"
#> [6,] "Average" "Average" "Disengaged" "Average" "Disengaged" "Disengaged"
#> 7 8
#> [1,] "Active" "Active"
#> [2,] "Average" "Average"
#> [3,] "Active" "Active"
#> [4,] "Average" "Active"
#> [5,] "Active" "Active"
#> [6,] "Average" "Average"
sort(unique(as.vector(trajectories)), na.last = NA)
#> [1] "Active" "Average" "Disengaged"sequence_plot() is the single entry point for three
views of this data:
type |
What it shows | Uses dendrogram? | Facets? |
|---|---|---|---|
"heatmap" (default) |
dense carpet, rows sorted by a distance/dendrogram | yes | no |
"index" |
carpet without dendrogram, row-gap optional | no | yes |
"distribution" |
stacked area / bar of state proportions over time | no | yes |
Defaults: legend = "right",
frame = FALSE.
type = "heatmap" — clustered carpetAvailable sorts: lcs (default), frequency,
start, end, plus any
build_clusters() distance — hamming,
osa, lv, dl, qgram,
cosine, jaccard, jw.
kCut the dendrogram into k groups and overlay thin
horizontal lines at the cluster boundaries in the ordered rows. Tune
with k_color and k_line_width.
sequence_plot(trajectories, k = 5,
k_color = "black", k_line_width = 1.2,
main = "k = 5 — thin black")sequence_plot(trajectories,
legend = "bottom",
legend_title = "Engagement",
state_colors = c("#2a9d8f", "#e9c46a", "#e76f51"),
main = "Custom palette + bottom legend")tick thinningtype = "index" — gap-ready carpet with facetsNo dendrogram. Rows are sorted within each panel by
sort. Supports group (vector or auto from a
net_clustering) plus ncol / nrow
facet grids.
type = "distribution" — state proportions over
timeStacked area or bar chart of state frequencies per time column.
sequence_plot(trajectories, type = "distribution",
geom = "bar", scale = "count",
main = "distribution — bars, count scale")# Always explore first with the default:
sequence_plot(trajectories)
# Zoom in on cluster structure:
sequence_plot(trajectories, k = 3)
sequence_plot(trajectories, sort = "hamming", k = 4)
# Compare cluster compositions:
cl <- build_clusters(as.data.frame(trajectories), k = 3,
dissimilarity = "hamming", method = "ward.D2")
sequence_plot(cl, type = "index")
sequence_plot(cl, type = "distribution")
# Polish for a paper:
sequence_plot(trajectories, k = 3,
state_colors = c("#2a9d8f", "#e9c46a", "#e76f51"),
legend_title = "Engagement",
legend = "bottom",
cell_border = "grey70",
main = "Student engagement trajectories")These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.