HSwls: Holzinger and Swineford (1939) partitioned into ordered...

In umx: Structural Equation Modeling and Twin Modeling in R

Holzinger and Swineford (1939) partitioned into ordered factors for WLS teaching.

Description

A dataset containing cognitive test scores x1 to x9 for 301 students from two schools. This is a modified version of the classic lavaan::HolzingerSwineford1939 or OpenMx's OpenMx::HS.ability.data dataset (which contains the original continuous data), where the variables x1 to x9 have been converted into ordered factors (2, 3, or 4 categories) using quantile-based cuts. It is designed for demonstrating and teaching Weighted Least Squares (WLS) estimation with ordinal data.

Usage

data(HSwls)

Format

A data frame with 301 rows and 15 variables.

Details

The variables are:

• id: Student ID

• sex: Gender (1 = male, 2 = female)

• ageyr: Age in years

• agemo: Age in months

• school: School (Grant-White or Pasteur)

• grade: Grade (7 or 8)

• x1: Visual perception (2-category ordered factor)

• x2: Cubes (2-category ordered factor)

• x3: Lozenges (2-category ordered factor)

• x4: Paragraph comprehension (3-category ordered factor)

• x5: Sentence completion (3-category ordered factor)

• x6: Word meaning (3-category ordered factor)

• x7: Speeded addition (4-category ordered factor)

• x8: Code (4-category ordered factor)

• x9: Speeded counting of dots (4-category ordered factor)

References

Holzinger, K. J., & Swineford, F. (1939). A study in factor analysis: The stability of a bi-factor solution. Supplementary Educational Monographs, no. 48. Chicago: University of Chicago Press.

See Also

OpenMx::HS.ability.data, umxRAM()

Other datasets: Anthro_LDSC, Fischbein_wt, GFF, Psych_LDSC, docData, iqdat, umx, us_skinfold_data

Examples

## Not run: 
data(HSwls)
# x1:x3 represent visual perception
# x4:x6 represent verbal ability
# x7:x9 represent speed

m1 = umxRAM("WLS_HS", data = HSwls, type = "DWLS",
  umxPath(v1m0 = c("visual", "verbal", "speed")),
  umxPath(from = "visual", to = c("x1", "x2", "x3")),
  umxPath(from = "verbal", to = c("x4", "x5", "x6")),
  umxPath(from = "speed" , to = c("x7", "x8", "x9")),

  umxPath(unique.bivariate = c("visual", "verbal", "speed")),
  umxPath(v.m. = c("x1", "x2", "x3", "x4", "x5", "x6", "x7", "x8", "x9"))
)
plot(m1, means=FALSE, std=TRUE, splines=FALSE, min = c("visual", "verbal", "speed"))
umxSummary(m1, uncertainty= "MLR")

data(HS.ability.data)
df = HS.ability.data

verbal = c("general" , "paragrap", "sentence")
space  = c("visual"  , "cubes"   , "flags"   )
math   = c("addition", "code"    , "counting")

manifests = c(verbal, space, math)
out = factanal(df[, manifests], factors = 3, scores = "Bartlett", na.action = na.exclude)

m1 = umxRAM("One Factor", data = df,
	umxPath(v1m0 = "G"),
	umxPath(v.m. = c(manifests, "V", "S", "M")),
	umxPath(from = "V", to = verbal),
	umxPath(from = "S", to = space),
	umxPath(from = "M", to = math),

	umxPath(from = "G", to = c("V", "S", "M"))
)
plot(m1, means=FALSE, std=TRUE, max= manifests, min="G", same= c("V", "S", "M"))

## End(Not run)

umx documentation built on Oct. 11, 2026, 5:07 p.m.