*2.3. Alternate Forms Evaluation*

The alternate forms reliability between the original HESPER scale and HESPER Web was evaluated using a voluntary convenience study sample of 50 study participants from the camp. Based on a power analysis that indicated a need for a minimum of 19 participants in both data collections in order to detect a statistically significant correlation and a power of 90%, and previous experiences from conducting alternative forms evaluation [11,14,15], a sample size of 50 was chosen. Inclusion criteria were that the person should be at least 18 years old, have access to the Internet by mobile phone, tablet or computer and be able to participate in the interview using the English language. For all participants, the HESPER interview was made prior to the web survey, due to practical reasons. The HESPER interviews were conducted by two male and two female volunteer assistants trained during a six-hour training session in accordance with the HESPER manual. Using a cluster random sampling method, four square areas within the Hagadera camp were first selected by lot to be included in the study [13]. Thereafter, the households asked to participate were selected using a kind of random walk method [8], where every second house in a direction pointed at by spinning a pencil was visited. In the first household, the first person to approach the interviewers was asked to participate. In the second household visited within the cluster, the second person seen by the interviewers was asked to participate, and so on. If any of the persons selected could not participate for any reason, the next person in the household was asked. The interviewers estimated that in every third household, there was no person eligible for participation. If so, the interviewers continued to the next household. A code list was used to group the HESPER scale and HESPER Web answers. Both data collections were answered anonymously, using the specific code only as reference in the web survey. The participants got a personally written reminder note from the interviewer, with the code and the link to HESPER Web, asking them to complete it within 48 h. The time between the HESPER interview and that taken to answer HESPER Web varied from a few hours up to three days.

#### *2.4. Feasibility Evaluation*

The sample for the feasibility evaluation of HESPER Web was conducted with 289 voluntary study participants who were recruited by advertising the study in the adult training centre, secondary schools and internet- and communication centers in Hagadera. Additionally, digital advertising on Facebook and on three specific pages aimed at people living in Hagadera or other Dadaab camps was used. Inclusion criteria were that the study participant should be at least 18 years old, have access to the Internet by mobile phone, tablet or computer and be able to participate in the survey using the English language. The

data collection period lasted for seven days (see Figure 1). The web survey was anonymous, and there were no limitations on answers from the same IP address, in order to allow several responders to use the same computer, tablet or mobile phone to answer the survey. Data were saved in a secured research database at Orebro University in Sweden.

**Figure 1.** Number of participants per day in the HESPER Web voluntary, self-recruited study sample.
