**3. Study 2**

## **3.1 Setting and participants**

Recruitment procedures and eligibility criteria for participation were the same as in Study 1. The personal growth group on sexual life met from January to June 2019 in the rehabilitation room of the USU-PG; it is accessible to wheelchairs. Only 7 participants were included in the present study, as explained in the subsection "Sample."

### **3.2 Measurements, procedures, and apparatus**

The self-administered sociodemographic questionnaire, outcome measures, procedures for administration, and structure, content, and techniques of the psychoeducational intervention were the same as in Study 1. A Sony ICD-PX312 audio recorder was added to the apparatuses described for Study 1 (Section 2.5) to record the dialogs of each meeting. The COnsolidated criteria for REporting Qualitative studies (COREQ ) checklist was followed in reporting the results [76]. The COREQ checklist is available at [77].

### **3.3 Data analysis**

*Outcome measures*. Nonparametric statistical tests were used due to the small sample size (Wilcoxon test for paired samples), and the *r* to determine the effect size of the intervention interpreted similarly to Cohen's *d* [75], as in Study 1.

*Qualitative data*. In grounded theory, data collection and analysis occur simultaneously [78]. The 12 meetings of the personal growth group were all audiotaped by two independent researchers (GC and ACM). All audio recordings from the 24 meetings were then transcribed, carefully read, and indexed according to the grounded theory coding procedure [78, 79]. To investigate the effects that the psychoeducational intervention has had on the participants' sexual health, the most significant verbal expressions were identified, and a code/index was assigned to each

#### *Psychological Sexual Health of People with Paraplegia DOI: http://dx.doi.org/10.5772/intechopen.91854*

one through a three-phase-inductive logic procedure, by studying the topic within its context and using an emerging design. See the Supplementary Material, where the codebook and detailed qualitative data analysis of each meeting are reported [77].

The reliability of codes was evaluated through measuring intercoder agreement, assessing the degree of trustworthiness of each code assigned to the same portion of text by GC and ACM. The level of agreement between two researchers was assessed using Krippendorff 's alpha for nominal items [80]; reliability was considered optimal if α ≥ 0.800, suboptimal with α ≥ 0.667, and non-optimal otherwise (**Table 1**).

The manual text analysis was transferred to Atlas.ti software (version 8.4) to measure intercoder agreement and calculate the occurrences of the categories and their relevance. Finally, through the software, the codes that emerged from the transcriptions of each meeting were compared to evaluate any differences in the theoretical construct that characterized them. By detecting how the codes developed with the progression of the meetings, it was possible to infer the effect of the psychoeducational intervention on the participants' sexual health.
