Figure 1
Data Collections in TwinLife.
Note: F2F: Face-to-face interview at household; CATI: Computer-assisted telephone interview, *The fourth face-to-face interview in the household had to be substituted with a CATI & CAWI (computer assisted web interview) due to the COVID-19 pandemic, in the fifth face-to-face data collections, participants can choose between a conventional household interview and a CATI & CAWI variant of the survey.
Figure 2
TwinLife COVID-19 Supplementary Surveys.
Note. F2F 1a – F2F 5b: First face-to-face interview for subsample A – Fifth face-to-face interview for subsample B; CATI 1a – CATI 4b: First computer-assisted telephone interview for subsample A – fourth computer-assisted telephone interview for subsample B; Saliva: collection of saliva via self-collection kit (Oragene kit, 2 ml); Cov 1 – Cov 4b: First COVID-19 supplementary survey – Fourth COVID-19 supplementary survey for subsample B; *The fourth face-to-face interview in the household had to be substituted with a CATI & CAWI (computer assisted web interview) due to the COVID-19 pandemic. In the fifth face-to-face data collection, participants can choose between a conventional household interview and a CATI & CAWI variant of the survey; 1. strict lockdown and 2. strict lockdown: contact restrictions in private life, travel restrictions, closure of schools, gastronomy, retail, service sector, cultural institutions (e.g., museums, cinema, concerts) and social sports activities; lighter lockdown measures: contact restrictions in private life and closure of gastronomy, cultural institutions and social sports activities.
Figure 3
Sample size development.
Note: Calculations are based on TwinLife Data file Version 6.1.0, https://doi.org/10.4232/1.13987 (F2F1 to F2F3) and preliminary data for recently collected data (CATI3, F2F4).
Table 1
Sample characteristics.
| DATA COLLECTIONS | |||||||
|---|---|---|---|---|---|---|---|
| F2F 1 | CATI 1 | F2F 2 | CATI 2 | F2F 3 | CATI 3 | F2F 4 | |
| Zygosity | |||||||
| Monozygotic | 1880 (46%) | 1332 (45%) | 1240 (45%) | 1050 (46%) | 1019 (45%) | 970 (45%) | 968 (45%) |
| Dizygotic | 2213 (54%) | 1603 (55%) | 1491 (55%) | 1249 (54%) | 1243 (55%) | 1185 (55%) | 1195 (55%) |
| Total | 4093 (100%) | 2935 (100%) | 2731 (100%) | 2299 (100%) | 2262 (100%) | 2155 (100%) | 2163 (100%) |
| Sex of twins | |||||||
| Male | 1854 (45%) | 1330 (45%) | 1257 (46%) | 1051 (46%) | 1037 (46%) | 977 (45%) | 992 (46%) |
| Female | 2242 (55%) | 1605 (55%) | 1475 (54%) | 1248 (54%) | 1225 (54%) | 1179 (55%) | 1171 (54%) |
| Total | 4096 (100%) | 2935 (100%) | 2732 (100%) | 2299 (100%) | 2262 (100%) | 2156 (100%) | 2163 (100%) |
| Migration background of twins | |||||||
| No migration background | 3074 (75%) | 2296 (78%) | 2155 (79%) | 1806 (79%) | 1799 (80%) | 1725 (80%) | 1734 (80%) |
| Migration background | 1022 (25%) | 639 (22%) | 577 (21%) | 493 (21%) | 463 (20%) | 431 (20%) | 429 (20%) |
| Total | 4096 (100%) | 2935 (100%) | 2732 (100%) | 2299 (100%) | 2262 (100%) | 2156 (100%) | 2163 (100%) |
| Mother’s highest education at F2F1 | |||||||
| Primary level (1) | 730 (18%) | 405 (14%) | 335 (12%) | 251 (11%) | 242 (11%) | 201 (9%) | 219 (10%) |
| Intermediate level (2) | 2031 (50%) | 1463 (50%) | 1328 (49%) | 1096 (48%) | 1090 (48%) | 1039 (48%) | 1039 (48%) |
| Higher level (3) | 1327 (32%) | 1065 (36%) | 1066 (39%) | 950 (41%) | 927 (41%) | 914 (42%) | 901 (42%) |
| Total | 4088 (100%) | 2933 (100%) | 2729 (100%) | 2297 (100%) | 2259 (100%) | 2154 (100%) | 2159 (100%) |
[i] Note: Calculations are based on TwinLife Data file Version 6.1.0, https://doi.org/10.4232/1.13987 (F2F1 to F2F3) and preliminary data for recently collected data (CATI3, F2F4). Migration background is assigned if either the twins were born abroad (first generation) or at least one parent was born abroad (second generation). Missing self-reports of the parents are filled in with the twins’ proxy information on the parents’ place of birth. In the case of completely missing information, it is assumed that there is no migration background. Mother’s education at F2F1 is based on CASMIN; missing information has been filled with father’s education.
Table 2
Overview of constructs in TwinLife with relevance for educational research.
| CONSTRUCT | DESCRIPTION | EXAMPLE ITEMS |
|---|---|---|
| Educational attainment | ||
| Information on education history and educational institutions | TwinLife assesses extensive information on education history and educational institutions including school attendance, type of school, grade level, skipping or repeating grade levels, recommendation for secondary school, desired school-leaving qualifications, after school-jobs, information on vocational training, and higher education | “Are you currently in education or training? In other words, are you attending school or higher education (including doctorate/Ph.D.), completing an apprenticeship or vocational training, or in further education or training?” |
| Tutoring and extracurricular activities | It is assessed whether children take part in extracurricular activities, including help with homework, remedial teaching, additional classes, and participation in clubs | “Does your child participate outside regular school hours in one or more of the following activities?” – Help with homework |
| Information on school grades and school report cards | Photographs of the report card were taken and coded by trained personnel (for more information, see Instinske et al., 2022). If there was no certificate available or the participants did not give their consent, questions about the school grades in maths and German were asked instead. For participants that have left school already, the grade point average was surveyed. | “Please indicate what school grade you had on your last report card in the following subjects”- Math/German |
| Cognitive abilities | The Culture Fair Test (CFT; Weiß, 2006; Weiß & Osterland, 2012) was applied. The test measures non-verbal (fluid) intelligence, which can serve as a proxy for general cognitive ability. For participants aged five to nine years, three different subtests (figural reasoning, figural classification, and matrices; CFT 1-R; Weiß & Osterland, 2012) were assessed. For participants ten years of age and older, four subtests were used (reasoning in addition to the aforementioned three subtests; CFT 20-R; Weiß, 2006). For more information, see Gottschling (2017). | See test manual of the CFT |
| Motivation and academic self-concept | ||
| Academic self-concept | Academic self-concept was assessed for school students with three items concerning their academic self-concept for school in general, but also three items specific for the subject maths and German. The assessment for school students was based on scales for academic self-concept (Dickhäuser et al., 2002; orig. “Skalen zum akademischen Selbstkonzept” (SESSKO)). | “I am [not talented – talented] for school” “I know [Just a little – A lot] in maths” |
| Intrinsic motivation | Intrinsic motivation was assessed for school students with three items concerning their intrinsic motivation for school in general, but also three items specific for the subject maths and German. Intrinsic motivation for school students was measured with adapted items of the scale for the assessment of subjective school values (Steinmayr & Spinath, 2010; orig. “Skala zur Erfassung subjektiver schulischer Werte” (SESSW)). | “I like doing the things I learn at school.” “Maths is fun.” |
| Learning motivation | The scales for the assessment of learning and performance motivation (Spinath et al., 2002); orig. “Skalen zur Erfassung der Lern- und Leistungsmotivation” (SELLMO-S) were adapted to measure learning motivation. The construct was assessed for school in general using three items. | “At school, I am interested in learning something interesting.” |
| Achievement motivation | Three items were developed for TwinLife. | “Good achievements mean a lot to me.” |
| Subjective perceptions of burden due to school and perception of the school climate | ||
| Feelings of burden due to school | Feelings of burden were assessed with an adapted version of questions originating from a survey by the German National Educational Panel Study (NEPS, 2012), called “additional study in Thuringia” (Blossfeld & Roßbach, 2019). It was assessed for all school students aged 13 or older with seven items. | “I often feel tense when I come home from school.” |
| Student-teacher interactions | This construct was assessed with five items, originating from the PISA survey (OECD, 2013). | “Students get along well with most teachers.” |
| Personality and other traits | ||
| Personality | Two different versions of the Big Five Inventory were used. Every participant over ten years of age rated their own personality on the Big Five Inventory – Short Version (BFI-S; Gerlitz & Schupp, 2005). The sub-scales conscientiousness, extraversion, agreeableness and neuroticism were assessed with three items each, while openness was assessed with four items. For younger children, their parents rated their personality on the Five factor questionnaire for children – short form (FFFK-K; Weinert et al., 2007); orig. “Fünf Faktoren Fragebogen für Kinder – Kurzform”, with 2 items for each subscale. | “I see myself as someone who does a thorough job.” “I see myself as someone who is talkative.” “My child is [Not that interested – Hungry for knowledge].” “My child is [Untidy – Tidy].” |
| Self-Regulation | Self-Regulation was assessed using items of two different questionnaires: Three items of the BISS scale, which is a German adaption of the Grit Scale (Consistency of Interests) from Fleckenstein et al., (2014); and three items of the German short version of the Self-Control Scale (SCS-K-D); Bertrams & Dickhäuser, (2009). For children aged nine or younger, only the Self-Control Scale was assessed via parental report on their children. | “New ideas and projects sometimes distract me from previous ones.” “I do certain things that are bad for me, if they are fun.” |
| Self-Eesteem | Depending on the age of children, self-esteem was assessed either as a self-report or as a parental report. As a self-report, self-esteem was assessed using three items from the “Panel Analysis of Intimate Relationships and Family Dynamics” (pairfam; see Thoennissen et al., 2014; items are based on the Rosenberg Self-Esteem Scale; RSE; Rosenberg, 1965). For the parental report (participants aged 5 to 12), two of these items were rephrased. | “I take a positive attitude toward myself.” “My child is self-confident.” |
| Self-Efficacy | Three items from the general self-efficacy short scale (ASKU; Beierlein et al., 2012; orig. “Allgemeine Selbstwirksamkeit Kurzskala”) were used to measure this construct. | “I can rely on my own abilities in difficult situations.” |
| Home environment and parental support | ||
| Quality of home environment | This construct was measured with an adapted version of the confusion, hubbub, and order scale (CHAOS; Matheny et al., 1995). It can be used to rate chaotic vs. orderly home environments. The scale consisted of six items. | “You can’t hear yourself think in our home.” |
| Parental involvement | The scales for parental involvement were inspired by the CoSMoS project (see Spinath & Wolf, 2006) and an instrument by Lorenz & Wild (2007). The items were assessed as a child report on their parents for four different facets with three items each: Structure, emotional support, autonomy and control. | “When I study for an exam, I know exactly how much effort my parents expect of me.” “When I get a poor grade, my parents complain and demand that I work harder.” |
| Parenting style | The scales for parenting style were inspired by pairfam (Huinink et al., 2011). Parenting style was assessed as parental self-report of each parent and child report on both their parents. There were 5 scales: Emotional warmth (four items), psychological control (three items), negative communication (two items), monitoring (two items), inconsistent parenting (two items) | “You show with words and gestures that you like your child.” “If your child does something against your will, you punish your child.” “Your mother/your father yells at you because you did something wrong.” |
| Family activities | Parents and children were asked how much time they spent together doing certain activities, like reading, singing or going to the playground in the last month. The questions were taken from the pairfam study. | “How often have your parents or other members of your family taken part in the following activities with you during the last four weeks?” – Singing and making music |
| Cultural capital | Depending on the age of the participant, cultural capital was assessed either as a self-report or as a parental report. The self-report items originated from the NEPS study (for more information regarding particular subscales, see Goßmann, 2018). Items for cultural capital included the categories embodied cultural capital (five items), cultural involvement (five items), and participation in high culture (five items). As a parental report, however, only participation in high culture was asked, with similar questions as in the self-report. Also, the participants were asked how much time they spent reading daily. | “Do you have books of poems at home” “In general, how often do you discuss the following things with others? – Political and social issues” |
| Social relationships | We collected data on the number of friends, age and sex of significant others, and also on the possible experience of loneliness. In addition, information was collected on sibling and parent-child relationships, support relationships, and information on conflicts and quarrels with others. | “How many close friends do you have? This includes family members who you are close to. This does not include Facebook friends.” “To what extent do you agree with the following statement? I often feel lonely.” “How satisfied are you with your relationship with your twin/sibling?” |
| COVID-19 pandemic | ||
| Pandemic related topics | In light of the COVID-19 pandemic, the TwinLife project decided to incorporate a special questionnaire series with COVID-19 pandemic specific questions to gain information on pandemic related changes in some of the core social inequality dimensions investigated in TwinLife: Burden and threats due to the COVID-19 pandemic; emotional stress and COVID-19 related coping and resilience; changes in employment and finances; worsening of pre-existing medical condition; changes in educational opportunities. | “During the Corona pandemic, schools were or are closed for extended periods of time. Are you attending classes normally right now?” “The Corona pandemic continues to affect the daily lives of many people. If you are currently experiencing the following limitations, how bad are they for you?” – Having schooling (also) at home |