Cristin-prosjekt-ID: 2472481
Sist endret: 7. september 2020, 13:27

Cristin-prosjekt-ID: 2472481
Sist endret: 7. september 2020, 13:27
Prosjekt

SMS versus E-mail for recruiting students to participate in a COVID-19 related survey: A Randomised Trial

prosjektleder

Arnfinn Helleve
ved Psykisk og fysisk helse ved Folkehelseinstituttet

prosjekteier / koordinerende forskningsansvarlig enhet

  • Folkehelseinstituttet

Tidsramme

Avsluttet
Start: 4. september 2020 Slutt: 31. oktober 2020

Beskrivelse Beskrivelse

Tittel

SMS versus E-mail for recruiting students to participate in a COVID-19 related survey: A Randomised Trial

Populærvitenskapelig sammendrag

Low response rates are a major challenge when conducting surveys. Using SMS to recruit respondents may be an effective tool, but is it worth the cost? In this project we will randomize 120,000 Norwegian university and college students to receive either SMS- or e-mail invitations to participate in a COVID-19 related survey. We will compare response rates after one week. We will also compare response rates after two weeks, and assess differences across gender and age groups.

Vitenskapelig sammendrag

Low response rates are a major challenge when conducting surveys. Using SMS to recruit respondents may be an effective tool, but comes at a cost. Whether the potential gain in response rate is worth the price, is unclear. To investigate this, we will randomize 120,000 Norwegian university and college students to receive either SMS- or e-mail invitations to participate in a COVID-19 related survey. All students will receive two reminders, so one group will receive three SMS invitations, and the other will receive three e-mail invitations. We will compare response rates after one week. We will also compare response rates after two weeks, and assess differences across gender and age groups.

Metode

The leadership at 13 higher learning institutions in Norway agreed to take part in a study to assess the association between in-person instruction and COVID-19 risk (the Campus & Corona study [6]). They were therefore willing to share the e-mail addresses and mobile phone numbers of their students with us. From a list with contact information for 125.491 students, we removed 5.557 individuals without an operating university e-mail account or a valid Norwegian cell phone number. We entered the contact information for the remaining 119.658 students into a spreadsheet, with one line per student. We used the random number generator integrated in R to give insert a unique random number to each student. We then rearranged the spreadsheet so that the students were by a random number, in rising order. We then allocated the first half to receive e-mail invitations, and the second half to receive SMS-invitations. Two of the authors (AH and RKH) conducted the randomization procedure.

We will send out all the invitations on Friday 4 September 2020. We will send two SMS-reminders to the SMS-group, and two e-mail reminders to the e-mail group. The first on Monday 7 September, and the second on Wednesday 9 September.

The e-mail contains information about the goal of the study and about a financial incentive, i.e. that those who take part in the study enter a lottery with five winners receiving a universal gift card of NOK 1000 (around $100). The SMS only contains a short invitation to participate, without further information about the study or the financial incentive. Both the e-mail and the SMS contain a link to the consent form for the study. Those who consent to participate are automatically directed to the web-based questionnaire. The consent form provides detailed information about the study.

Both the SMS and the e-mail invitations contain a link to the survey that the recipients are encouraged to click.

Outcomes and analyses

We will compare response rates (measured as completed consent forms or completed questionnaires) one week after the first dispatch of invitations, and calculate the relative response rate with 95% CI. This serves as our primary outcome.

We will also calculate the relative response rate after two weeks. In addition, we will conduct sub group analyses based on gender, age groups, and possibly other background variables depending on availability (e.g. field of study of sociodemographic). We may also explore differences in proportions of invitees who open the informed consent form, and the proportions who enter the questionnaire itself.

We will present response rates over time (after first invitation, first and second follow-up) graphically.

Tittel

SMS mot e-post for rekruttere studenter til å delta i en spørreskjemaundersøkelse om covid-19: et randomisert forsøk

Populærvitenskapelig sammendrag

Lav responsrate er et kjent problem ved spørreskjemaundersøkelser. Bruk av SMS kan være en effektiv måte å rekruttere deltakere på, men er det verdt prisen? I dette prosjektet vil vil fordele 120 000 studenter i Norge til to grupper ved hjelp av loddtrekning. Disse vil motta invitasjon til å delta i en spørreskjemaundersøkelse, ved henholdsvis SMS eller epost. Vi vil sammenlikne responsen etter ei uke. Vi vil også sammenlikne responsraten etter to uker, og vi vil undersøke forskjeller på tvers av kjønn og aldersgrupper.

prosjektdeltakere

prosjektleder

Arnfinn Helleve

  • Tilknyttet:
    Prosjektleder
    ved Psykisk og fysisk helse ved Folkehelseinstituttet

Rannveig Kaldager Hart

  • Tilknyttet:
    Prosjektdeltaker
    ved Psykisk og fysisk helse ved Folkehelseinstituttet
Aktiv cristin-person

Atle Fretheim

  • Tilknyttet:
    Prosjektdeltaker
    ved Helsetjenester ved Folkehelseinstituttet

Dagfinn Bergsager

  • Tilknyttet:
    Prosjektdeltaker
    ved Universitetets senter for informasjonsteknologi ved Universitetet i Oslo
1 - 4 av 4