3.1. Study Design and Setting
This one group quasi-experimental with pre- and post-test study was conducted from August to December 2014, in the Iranian epilepsy association. The subjects were selected by convenience sampling. The sample size included 40 children with epilepsy. Inclusion criteria included: diagnosis of epilepsy and idiopathic epileptic syndrome established according to the clinician neurologist diagnostic examination, age 6 - 12 years, with a regular school attendance and living in Hamadan city. Exclusion criteria were a positive history of mental retardation, psychiatric disorders, and chronic physical diseases of child as well as a positive family history of any diagnostic psychiatric disorders in parents and sibling.
The sample size in this study was calculated according to the Rahmati et al. (2010) study (
13).

Equation 1.

Equation 2.
3.2. Instruments
In this study, 2 questionnaires included demographic and social skills rating system (SSRS) were used to collect data. Demographic questionnaire included children’s personal variables such as age, gender, birth order, education, and parents personal variables such as age, gender, educational degree, marital status, occupation, history of mental illness, history of psychiatric medication use or any other diseases. Social skills rating system was used for assessing social skills of children with epilepsy. The social skills rating system (Gresham and Elliot 1990) is a norm-referenced assessment tool that focuses on social skills and problem behaviors in pre-school, elementary, and secondary students (
14). The SSRS consists of 3 forms for students, parents, and teachers (
15). The SSRS focuses on the following measurement areas: (1) social skills (2) problem behaviors and (3) academic competence (in the teacher report form). Each questionnaire contains 34 to 57 items on a 3-point scale (0 = never, 1 = sometimes, 2 = very often) that assess the individual’s social behavior. We used the parent form. The parent form asks the parent or caregiver to rate the behavior of the child to describe the individual’s typical behavior.
The Social Skills scale has 5 subscales: cooperation, assertion, responsibility, empathy, and self-control. The sum of the subscales yield “behavior levels” that indicate the descriptive frequency of a particular behavior. All subscale scores can be summed to provide standard scores and percentile ranks based on the normative sample.
Since its release, there is much of the published evidence of suitable reliability and validity for the SSRS has focused primarily on the teacher and parent report forms (
16,
17). Internal consistency yielded average coefficient alpha reliabilities of 0.90 for the Social Skills scale. Test-retest reliability for parent scores of the elementary-aged was 0.80 on the Social Skills scale (
16).
Reliability of the parent form of the questionnaire were assessed in Iranian children and Cronbach’s alpha α = 0.87 showed appropriate internal consistency of the social skills rating system (
18,
19). Questionnaires were completed as self-report by one of the parents.
3.3. Data Collection and Intervention Methods
First, the time and place of classes were organized with the participants. The educational program was composed of 4 sessions and each session would last for 1 hour. Before the first session, parents learned about educational interventions and methods of completing questionnaires. They were then asked to complete the questionnaires before and 1 month after intervention.
Social skills training have been widely used as interventions with children and adolescents in order for them to attain social competence. Social skills training programs are designed to teach pro-social skills and social adjustment. Social skills training programs are designed to allow an individual to attain 3 goals: (a) development of positive interpersonal relationships, (b) ability to cope with expectations of various social situations, and (c) execution of effective communication in social situations. Contends that most contemporary approaches have been divided into 4 approaches: (a) the social skills approach, (b) the social problem-solving approach, (c) the social perspective taking, and (d) the self-control training (
20).
The sessions in the present study were focused on the following issues:
First session was orientation. The children and their patents introduced them and shared their experiences. Necessary information regarding epilepsy and its influence of epilepsy on lifestyle of children and their family were provide based on their needs. Then, volunteer children and parents explained one of his/her social experience with someone else such as their sister, brother, friends, peers, or relatives. Therefore the current social skills approach of them was tested.
In the second session, social skills approach and the effects on life and the social problem-solving approach were taught. For example, friendly relationships, selecting a good friend, fulfill their expectations, keeping their relationships, what to expecting from their friends, and so on.
Third session was the social perspective taking, such as cooperation in the family, among a group of peers and their friends. For this purpose, cooperation was defined and the good points of cooperating with other people were explained. Then, the children were asked to draw pictures together or take part in a group playing.
Fourth session was the self-control training, especially in anger and wrath. First the children were asked explain their definition, how they express their anger and ultimately, their using strategies to control their anger. Then, some information was taught about self-control techniques such as deep breathing and using alternative methods to express anger like fisting into pillows. At the end of the sessions, pamphlets and educational booklets about the topics with more details were given to the participants.
3.4. Data Analysis
The data was analyzed using SPSS 16, with descriptive statistics including frequency; mean and standard deviation, Kolmogorov-Smirnov test, and t-test for comparing the mean score before and after intervention. The significance level was considered less than 0.05.