The rSC function generates random single-case data frames
for monte-carlo studies and demonstration purposes.
design_rSC is used to set up a design matrix with all parameters needed for the rSC function.
rSC(design = NULL, round = NA, random.names = FALSE, seed = NULL, ...)design_rSC(
n = 1,
phase.design = list(A = 5, B = 15),
trend = list(0),
level = list(0),
slope = list(0),
rtt = list(0.8),
m = list(50),
s = list(10),
extreme.p = list(0),
extreme.d = c(-4, -3),
missing.p = list(0),
distribution = "normal",
prob = 0.5,
MT = NULL,
B.start = NULL
)
A design matrix which is created by design_rSC and specifies all paramters.
Rounds the scores to the defined decimal. To round to the second
decimal, set round = 2.
Is FALSE by default. If set random.names =
TRUE cases are assigned random first names. If set "male" or
"female" only male or female names are chosen. The names are drawn from
the 2,000 most popular names for newborns in 2012 in the U.S. (1,000 male
and 1,000 female names).
A seed number for the random generator.
Paramteres that are directly passed from the rSC function to the design_rSC function for a more concise coding.
Number of cases to be created (Default is n = 1).
A vector defining the length and label of each phase.
E.g., phase.length = c(A1 = 10, B1 = 10, A2 = 10, B2 = 10).
Defines the effect size d of a trend per MT added
across the whole data-set. To assign different trends to several
single-cases, use a vector of values (e.g. trend = c(.1, .3, .5)).
If the number of cases exceeds the length of the vector, values are
repeated. While using a binomial or poisson distribution, d.trend
indicates an increase in points / counts per MT.
Defines the level increase (effect size d) at the
beginning of phase B. To assign different level effects to several
single-cases, use a vector of values (e.g. level = c(.2, .4, .6)).
If the number of cases exceeds the length of the vector, values are
repeated. While using a binomial or poisson distribution, level
indicates an increase in points / counts with the onset of the B-phase.
Defines the increase in scores - starting with phase B -
expressed as effect size d per MT. slope = .1 generates an
incremental increase of 0.1 standard deviations per MT for all phase B
measurements. To assign different slope effects to several single-cases,
use a vector of values (e.g. slope = c(.1, .2, .3)). If the number
of cases exceeds the length of the vector, values are repeated. While using
a binomial or poisson distribution, d.slope indicates an increase in
points / counts per MT.
Reliability of the underlying simulated measurements. Set
rtt = .8 by default. To assign different reliabilities to several
single-cases, use a vector of values (e.g. rtt = c(.6, .7, .8)). If
the number of cases exceeds the length of the vector, values are repeated.
rtt has no effect when you're using binomial or poisson distributed
scores.
Mean of the sample distribution the scores are drawn from. Default
is m = 50. To assign different means to several single-cases, use a
vector of values (e.g. m = c(50, 42, 56)). If the number of cases
exceeds the length of the vector, values are repeated.
Standard deviation of the sample distribution the scores are drawn
from. Set to s = 10 by default. To assign different variances to
several single-cases, use a vector of values (e.g. s = c(5, 10,
15)). If the number of cases exceeds the length of the vector, values are
repeated.
Probability of extreme values. extreme.p = .05 gives
a five percent probability of an extreme value. A vector of values assigns
different probabilities to multiple cases. If the number of cases exceeds
the length of the vector, values are repeated.
Range for extreme values, expressed as effect size d.
extreme.d = c(-7,-6) uses extreme values within a range of -7 and -6
standard deviations. In case of a binomial or poisson distribution,
extreme.d indicates points / counts. Caution: the first value must
be smaller than the second, otherwise the procedure will fail.
Portion of missing values. missing.p = 0.1 creates
10% of all values as missing). A vector of values assigns different
probabilities to multiple cases. If the number of cases exceeds the length
of the vector, values are repeated.
Distribution of the scores. Default is distribution
= "normal". Possible values are "normal", "binomial", and
"poisson". If set to "normal", the sample of scores will be
normally distributed with the parameters m and s as mean and
standard deviation of the sample, including a measurement error defined by
rtt. If set to "binomial", data are drawn from a binomial
distribution with the expectation value m. This setting is useful
for generating criterial data like correct answers in a test. If set to
"poisson", data are drawn from a poisson distribution, which is very
common for count-data like behavioral observations. There's no measurement
error is included. m defines the expectation value of the poisson
distribution, lambda.
If distribution (see below) is set "binomial",
prob passes the probability of occurrence.
Number of measurements (in each study). Default is MT = 20.
Phase B starting point. The default setting B.start = 6
would assign the first five scores (of each case) to phase A, and all
following scores to phase B. To assign different starting points for a set
of multiple single-cases, use a vector of starting values (e.g.
B.start = c(6, 7, 8)). If the number of cases exceeds the length of
the vector, values will be repeated.
A single-case data frame. See scdf to learn about this format.
# NOT RUN {
## Create random single-case data and inspect it
design <- design_rSC(
n = 3, rtt = 0.75, slope = 0.1, extreme.p = 0.1,
missing.p = 0.1
)
dat <- rSC(design, round = 1, random.names = TRUE, seed = 123)
describeSC(dat)
plotSC(dat)
## And now have a look at poisson-distributed data
design <- design_rSC(
n = 3, B.start = c(6, 10, 14), MT = c(12, 20, 22), m = 10,
distribution = "poisson", level = -5, missing.p = 0.1
)
dat <- rSC(design, seed = 1234)
pand(dat, decreasing = TRUE, correction = FALSE)
# }
Run the code above in your browser using DataLab