14  Exporting scan results

The export function call

export(object, …)

The export function will make it easier to convert the results of your scan analyses into tables and descriptions you can add to your documents and presentations. Basically, export takes a scan object and converts it to an html-table or latex output.

Note

export has two “engines” implemented to create the output formats. The “kable” engine is build on top of the knitr and kableextra packages. The list provided in the kable_options argument is implemented in the kable function of knitr and the list provided to the kable_styling_options is implemented in the kable_styling command of the kableExtra package. export sets some defaults for these functions but you can play around and overwrite them.

The “gt” engine is a more recent development and build ontop of the gt package. “gt” allows to export to html, rtf, latex, docx, pdf, and png.

You can change the use of the engine by setting options("scan.export.engine" = "gt") or options("scan.export.engine" = "kable").

export works best when used within an rmarkdown file and/or within RStudio. In RStudio, the html table will be displayed in the Viewer pane. There you can click the export button () to export an html or bitmap file or you can try drag and drop ➡️ copy and paste the table into another application.

Alternatively, you can set the filename argument to export the table directly from within the export function. The file name extension you provide will define the resulting file format (e.g. filename = "results.html").

If you use the “kable engine”, possible extensions and file formats are html, png, and jpg. The newer “gt engine” also allows to export to docx, rtf, pdf, and tex formats. To apply the new eninge, set options("scan.export.engine" = "gt").

The examples provided in this book are created with the gt engine.

exampleAB |> 
  overlap() |> 
  export("my_file.html")

Most of the tables will have a default caption and footnotes. If you want to change these, use the caption and footnote arguments.

14.1 Single case data files

The scdf export function call

export(object, summary = FALSE, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), cols, round = 2, …)

export(exampleA1B1A2B2_zvt)

Single case data frame with three cases
Tick				Trick				Track
zvt	d2	day	part	zvt	d2	day	part	zvt	d2	day	part
47	131	1	A1	51	100	1	A1	54	89	1	A1
58	134	2	A1	58	126	2	A1	57	116	2	A1
76	141	3	A1	70	130	3	A1	51	114	3	A1
63	141	4	B1	65	130	4	B1	61	131	4	B1
71	140	5	B1	67	137	5	B1	57	132	5	B1
59	140	6	B1	63	133	6	B1	53	130	6	B1
64	138	7	A2	64	136	7	A2	58	128	7	A2
69	140	8	A2	70	137	8	A2	57	131	8	A2
72	141	9	A2	70	135	9	A2	60	130	9	A2
77	140	10	B2	68	128	10	B2	55	129	10	B2
76	138	11	B2	69	137	11	B2	58	118	11	B2
73	140	12	B2	70	138	12	B2	58	131	12	B2

14.2 Descriptive stats

The describe export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), flip = FALSE, round = 2, …)

The flip argument will rotate the table by 90 degrees.

res <- describe(GruenkeWilbert2014)
export(res)

Descriptive statistics
Case	Design	n		Missing		M		Median		SD		MAD		Min		Max		Trend
		A	B	A	B	A	B	A	B	A	B	A	B	A	B	A	B	A	B
Anton	A-B	4	14	0	0	5.00	9.14	5	9	0.82	0.77	0.74	1.48	4	8	6	10	-0.40	0.03
Bob	A-B	7	11	0	0	3.00	8.82	3	9	0.82	0.87	1.48	0.00	2	7	4	10	0.04	0.04
Paul	A-B	6	12	0	0	3.83	8.83	4	9	0.75	0.72	0.74	0.74	3	8	5	10	-0.26	0.02
Robert	A-B	8	10	0	0	4.12	8.90	4	9	0.83	0.99	1.48	1.48	3	7	5	10	-0.06	-0.14
Sam	A-B	5	13	0	0	4.60	9.08	5	9	0.55	0.86	0.00	1.48	4	8	5	10	0.10	0.03
Tim	A-B	4	14	0	0	3.00	9.00	3	9	0.82	0.96	0.74	1.48	2	7	4	10	-0.60	0.00
Note. n = Number of measurements; Missing = Number of missing values; M = Mean; Median = Median; SD = Standard deviation; MAD = Median average deviation; Min = Minimum; Max = Maximum; Trend = Slope of dependent variable regressed on measurement-time.

14.3 Standardized mean differences

The smd export function call

export(object, caption = NA, footnote = NA, filename = NA, select = c(“Case”, Mean A = “mA”, Mean B = “mB”, SD A = “sdA”, SD B = “sdB”, SD Cohen = “sd cohen”, SD Hedges = “sd hedges”, “Glass’ delta”, “Hedges’ g”, “Hedges’ g correction”, “Hedges’ g durlak correction”, “Cohen’s d”), kable_styling_options = list(), kable_options = list(), round = 2, decimals = 2, flip = FALSE, …)

The flip argument will rotate the table by 90 degrees.

smd(exampleAB) |> export(flip = TRUE)

Standardizes mean differences. Comparing phase 1 against phase 2
Statistic	Johanna	Karolina	Anja
Mean A	54.60	51.80	53.60
Mean B	74.13	73.47	74.07
SD A	2.41	6.83	3.05
SD B	8.94	9.76	7.57
SD Cohen	6.55	8.43	5.77
SD Hedges	7.97	9.19	6.83
Glass' delta	8.11	3.17	6.71
Hedges' g	2.45	2.36	3.00
Hedges' g correction	2.35	2.26	2.87
Hedges' g durlak correction	2.23	2.14	2.72
Cohen's d	2.98	2.57	3.55
Note. SD Cohen = unweigted average of the variance of both phases; SD Hedges = weighted average of the variance of both phases with a degrees of freedom correction; Glass’ delta = mean difference divided by the standard deviation of the A-phase; Hedges’ g = mean difference divided by SD Hedges; Hedges’ g (durlak) correction = approaches for correcting Hedges’ g for small sample sizes; Cohens d = mean difference divided by SD Cohen.

14.4 Trend analysis

The trend export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), round = 2, decimals = 2, …)

exampleABC$Marie %>%
  trend() %>%
  export()

Trend analysis
Phase	Intercept	B	Beta
values ~ mt
Linear.ALL	55.16	0.61	0.39
Linear.A	60.62	−1.92	−0.70
Linear.B	74.85	−0.61	−0.16
Linear.C	68.87	−0.19	−0.05
values ~ I(mt^2)
Quadratic.ALL	59.13	0.02	0.33
Quadratic.A	57.94	−0.21	−0.71
Quadratic.B	73.22	−0.04	−0.10
Quadratic.C	68.49	−0.02	−0.04
Note. Measurement-times start at 0 for each phase.

14.5 Overlap indices

The overlap export function call

export(object, caption = NA, footnote = NULL, filename = NA, kable_styling_options = list(), kable_options = list(), round = 2, decimals = 2, flip = FALSE, …)

The flip argument will rotate the table by 90 degrees.

exampleA1B1A2B2_zvt %>%
  select_phases(A = c(1,3), B = c(2,4)) %>%
  overlap() %>%
  export(flip = TRUE)

Overlap indices. Comparing phase 1 against phase 2
Statistic	Tick	Trick	Track
PND	16.67	0.00	16.67
PEM	66.67	50.00	50.00
PET	66.67	33.33	33.33
NAP	68.06	51.39	58.33
NAP-R	36.11	2.78	16.67
PAND	66.67	50.00	66.67
IRD	0.33	0.33	0.17
Tau-U (A + B - trend A)	0.07	−0.16	−0.04
Tau-U (A + B - trend A + trend B)	0.14	0.03	−0.03
Base Tau	0.27	−0.25	0.13
Delta M	5.50	3.17	0.83
Delta Trend	−0.31	−1.10	−0.74
SMD	0.52	0.40	0.26
Hedges g	0.56	0.50	0.26
Note. PND = Percentage Non-Overlapping Data; PEM = Percentage Exceeding the Median; PET = Percentage Exceeding the Trend; NAP = Nonoverlap of all pairs; NAP-R = NAP rescaled; PAND = Percentage all nonoverlapping data; IRD = Improvement rate difference; Tau U (A + B - trend A) = Parker’s Tau-U; Tau U (A + B - trend A + trend B) = Parker’s Tau-U; Base Tau = Baseline corrected Tau; Delta M = Mean difference between phases; Delta Trend = Trend difference between phases; SMD = Standardized Mean Difference; Hedges g = Corrected SMD.

14.6 Percentage non-overlapping data

The pnd export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), select = c(“Case”, “PND”, “Total”, “Exceeds”), round = 2, …)

pnd(Huber2014) |> export()

Percentage Non-Overlapping Data
Case	PND	Total	Exceeds
Adam	0.00	27	0
Berta	0.00	19	0
Christian	15.15	66	10
David	65.15	66	43

14.7 Percentage exceeding the trend

The pet export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), round = 1, …)

pet(exampleAB_decreasing) |> export()

Percent Exceeding the trend
Case	PET	Percentage > upper 95 CI	p (binomial test)
Peter	15.4	0	.99
Tony	0.0	0	1.00
Bruce	64.3	0	.21
Note. Assumed increasing values in the B-phase; Binomial test alternative hypothesis: true probability > 50%; Single-sided test.

14.8 Percentage exceeding the median

The pem export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), round = 2, …)

pem(Huber2014) |> export()

Percent Exceeding the Median
Case	PEM	positives	total	binom.p
Adam	25.93	7	27	1
Berta	15.79	3	19	1
Christian	98.48	65	66	0
David	96.97	64	66	0

14.9 Nonoverlap of all pairs

The nap export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), select = c(“Case”, “NAP”, “NAP Rescaled”, “w”, “p”, “d”, “R²”), round = 2, …)

nap(Huber2014) |> export()

Nonoverlap of all pairs
Case	NAP	NAP Rescaled	w	p	d	R²
Adam	28.15	-43.70	194.0	.97	-0.69	0.11
Berta	17.11	-65.79	157.5	.99	-1.00	0.20
Christian	86.97	73.94	86.0	<.001	1.70	0.42
David	91.74	83.48	54.5	<.001	2.06	0.51

14.10 Tau-U

The tauu export function call

export(object, caption = NA, footnote = NA, filename = NA, select = “auto”, kable_styling_options = list(), kable_options = list(), meta = FALSE, round = 3, decimals = 3, …)

Set the argument meta = TRUE to get the results of the meta analysis or set meta = FALSE (the default) to get a table with each case.

tau_u(exampleAB_decreasing) |> export()

Tau-U analyses
Model	Tau	CI lower	CI upper	Z	p
Peter
A vs. B	−0.978	−0.991	−0.944	−3.530	<.001
Trend A	−0.293	−0.857	0.590	−0.911	.36
Trend B	−0.104	−0.619	0.474	−0.490	.62
A vs. B - Trend A	−0.574	−0.810	−0.176	−3.184	<.01
A vs. B + Trend B	−0.546	−0.796	−0.136	−3.228	<.01
A vs. B + Trend B - Trend A	−0.483	−0.762	−0.051	−2.957	<.01
Tony
A vs. B	−0.979	−0.992	−0.947	−3.630	<.001
Trend A	−0.182	−0.786	0.600	−0.623	.53
Trend B	−0.092	−0.633	0.509	−0.413	.67
A vs. B - Trend A	−0.584	−0.816	−0.191	−3.282	<.01
A vs. B + Trend B	−0.575	−0.811	−0.177	−3.367	<.001
A vs. B + Trend B - Trend A	−0.504	−0.774	−0.079	−3.087	<.01
Bruce
A vs. B	−0.976	−0.991	−0.940	−3.383	<.001
Trend A	0.067	−0.788	0.833	0.188	.85
Trend B	−0.331	−0.733	0.242	−1.645	.10
A vs. B - Trend A	−0.607	−0.827	−0.225	−3.345	<.001
A vs. B + Trend B	−0.616	−0.832	−0.238	−3.691	<.001
A vs. B + Trend B - Trend A	−0.596	−0.822	−0.209	−3.668	<.001
Note. Method is ’ complete ‘. Analyses based on Kendall’s Tau b . 95 % CIs for tau are reported.

tau_u(exampleAB_decreasing) |> export(meta = TRUE)

Overall Tau-U
Model	Tau U	se	CI lower	CI upper	z	p
A vs. B	−0.978	0.140	−0.987	−0.962	−16.036	<.001
A vs. B - Trend A	−0.588	0.140	−0.740	−0.381	−4.822	<.001
A vs. B + Trend B	−0.579	0.140	−0.733	−0.369	−4.725	<.001
A vs. B + Trend B - Trend A	−0.530	0.140	−0.698	−0.305	−4.210	<.001
Note. Method is ’ complete ‘. Analyses based on Kendall’s Tau b . 95 % CIs for tau are reported.

14.11 Piecewise linear models

The plm export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), nice = TRUE, …)

res <- plm(exampleA1B1A2B2$Pawel)
export(res)

Piecewise-regression model predicting variable ‘values’
Parameter	B	CI(95%)		SE	t	p	delta R²
		2.5%	97.5%
Intercept	12.691	6.184	19.198	3.320	3.823	<.01
Trend mt	0.224	-0.995	1.443	0.622	0.361	.72	0.001
Level phase B1	16.285	6.308	26.261	5.090	3.199	<.01	0.118
Level phase A2	1.479	-18.806	21.764	10.350	0.143	.88	0.000
Level phase B2	11.455	-20.492	43.401	16.299	0.703	.48	0.006
Slope phase B1	-1.406	-3.130	0.318	0.879	-1.599	.12	0.029
Slope phase A2	-1.103	-2.827	0.621	0.879	-1.254	.21	0.018
Slope phase B2	-1.085	-2.809	0.639	0.879	-1.234	.22	0.017
Note. F(7, 32) = 7.86; p <.001; R² = 0.632; Adjusted R² = 0.552.

14.12 Hierarchical piecewise regressions

The hplm export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), round = 2, nice = TRUE, casewise = FALSE, …)

exampleAB_50 %>%
  add_l2(exampleAB_50.l2) %>%
  hplm(lr.test = TRUE, random.slopes = TRUE) %>%
  export()

Hierarchical Piecewise Linear Regression predicting variable ‘values’
Predictors	B	SE	df	t	p
Fixed effects
Intercept	48.21	1.4	1328	34.5	<.001
Trend mt	0.62	0.11	1328	5.52	<.001
Level phase B	13.87	0.89	1328	15.51	<.001
Slope phase B	0.86	0.12	1328	7.43	<.001
Random effects
	SD	L	df	p
Intercept	9.35	348.85	4	<.001
Trend mt	0.1	0.83	4	.93
Level phase B	4.54	42.82	4	<.001
Slope phase B	0.13	0.76	4	.94
Residual	4.97	NA	NA	NA
Model
AIC	8693.2
BIC	8771.7
ICC	0.29	L = 339	p <.001
Note. Estimation method ML; Slope estimation method: W; first; 50 cases.

14.13 Power analyses

The power_test export function call

export(object, caption = NA, footnote = NA, filename = NA, kable_styling_options = list(), kable_options = list(), round = 3, …)

design <- design(
  n = 1, phase_design = list(A = 6, B = 9),
  rtt = 0.8, level = 1.4, trend = 0.05
)
set.seed(124)
power_test(design, n_sim = 10) |> export()

Test power in percent
Method	Power	Alpha Error	Alpha:Beta	Correct
plm_level	80	10	1:2.0	85
rand	90	30	1:0.3	80
tauU	100	10	1:0.0	95