@mastersthesis{bal2025literary,
  author = {Bal, Onur},
  title = {Literary Work Readership and Its Relationship with Mentalisation and Empathy},
  school = {Yeditepe University},
  year = {2025},
  type = {Master's thesis},
  url = {https://onurbal.com/en/academics/thesis/}
}

4.1. Descriptive Analyses

Descriptive statistics were computed for book count and age, as well as for reading behaviour variables that were originally ordinal but treated as approximately continuous for summarisation purposes. Table 2 presents the minimum, maximum, mean, standard deviation, skewness, and kurtosis values for each of the demographic and reading-related variables.

Variable	Min	Max	M	SD	Skewness	Kurtosis
Age	19	30	22.1	2.3	1.3	1.4
Books	0	68	13	10.1	1.9	5.7
Textbook	0	7	3	2.1	0.5	-0.7
Academic	0	7	2	1.7	1.2	1.2
Magazine	0	7	0	1.2	2.4	6.9
Newspaper	0	7	0	0.9	3.1	13.4
E-Mail	0	7	1	1.1	3.2	13.6
Internet	0	7	4	2.2	-0.3	-1.1

Table 2\ Descriptive statistics of age, book count, and reading habits

Note. Age is in years; Books refers to the total number of books read from the book list (out of 101). Reading time variables were assessed using 8 ordered categories (“0 hours” to “7 or more hours”); means and standard deviations reflect approximate midpoints for descriptive purposes only.

According to widely cited guidelines, univariate distributions can be regarded as acceptably normal when skewness is between ±2 and kurtosis between ±7 (West, Finch, & Curran, 1995); more stringent conventions treat values within ±1 for skewness and ±2 for kurtosis as strongly normal (George & Mallery, 2010).

Among the variables examined, textbook reading time and internet-text reading time approximated normality under both the regular and stricter criteria. Book count, age, and reading time for academic texts met the regular thresholds but not the stricter ones. In contrast, reading time for magazines, newspapers, and e-mails showed substantial right skew and high kurtosis, indicating heavy tails and a clustering of values near the lower end of the scale. The distributions of these variables can be better understood by consulting Figures 1, 2, and 3. Figure 1 illustrates the age distribution of the participants, while Figure 2 shows the distribution of the number of books read. Figure 3 presents the distributions of weekly reading time across various formats, including textbooks, academic papers, magazines, newspapers, e-mails, and internet content.

Note. The overlaid curve is a kernel density estimate (KDE), providing a smoothed approximation of the age distribution.

*Figure 2* Distribution of the number of books read by the participants

Note. The overlaid curve is a kernel density estimate (KDE), providing a smoothed approximation of the book count distribution.

Although the age and book count variables met conventional thresholds for acceptable normality, they exceeded stricter criteria for strong normality. Log transformation was therefore applied to better satisfy the linearity assumption and reduce the influence of skew. An even more important rationale, however, was theoretical: changes at lower levels (e.g., reading 3 versus 6 books, or being 19 versus 23 years old) may carry greater psychological or developmental significance than equivalent increases at higher levels (e.g., 20 versus 23 books, or 24 versus 28 years old). The transformation thus also served to reflect the non-linear nature of these relationships in subsequent analyses.

Descriptive statistics were also computed for the total and subscale scores of the psychological instruments used in the study. These values are presented in Table 3.

Scale	Subscale	Component	Min	Max	M	SD	Skew.	Kurt.
STAI–S			20	77	41.8	12.4	0.36	-0.46
STAI–T			29	73	48.4	9.0	0.17	-0.38
ECQ			22	108	61.9	18.7	-0.02	-0.77
	General EA		13	64	37.7	11.9	-0.02	-0.65
	Death		5	25	14.5	5.3	0.16	-0.89
	Avoidance		4	20	9.7	3.8	0.43	-0.55
MentS			53	125	96.8	12.3	-0.39	0.04
	Self		8	40	27.2	6.7	-0.50	-0.31
	Other		19	45	36.8	4.9	-0.52	0.15
	Motivation		12	40	32.8	4.8	-0.84	1.21
QCAE			41	120	98.1	11.7	-0.75	1.39
	Cognitive		24	76	60.8	8.0	-0.56	0.85
		Perspective Taking	12	40	33.4	5.0	-0.68	0.34
		Online Simulation	9	36	27.4	4.4	-0.44	0.63
	Affective		17	47	37.3	5.4	-0.53	0.31
		Emotion Contagion	4	16	12.4	2.6	-0.56	0.10
		Proximal Responsivity	4	16	12.6	2.3	-0.64	0.47
		Peripheral Responsivity	5	16	12.2	2.2	-0.23	-0.54

Table 3\ Descriptive statistics of questionnaire scores

Note. Skew = Skewness, Kurt = Kurtosis.

All questionnaire scale and subscale scores comfortably met even the stricter normality criteria, with absolute skewness values not exceeding 0.84 and absolute kurtosis values remaining below 1.39. Combined with their strong internal consistency (as reported in the Methods section), the normal distributions of these scale scores suggest that the measures are both statistically reliable and appropriate for subsequent parametric analyses.