| Name | Speichervolumen | Metadaten | Upload | Aktion |
|---|
Alternativer Identifier:
-
Ersteller/in:
Bollenbach, Lukas
https://orcid.org/0000-0002-0321-0340
[Universität Konstanz]
Beitragende:
(Researcher)
(Researcher)
(Project Leader)
Niermann, Christina
https://orcid.org/0000-0002-2087-5328
[Medical School Hamburg]
(Researcher)
Schmitz, Julian
https://orcid.org/0000-0002-8725-4884
[Institut für Landes- und Stadtentwicklungsforschung gGmbH]
(Project Leader)
Kanning, Martina
https://orcid.org/0000-0001-6471-6268
[Universität Konstanz]
Titel:
cross-sectional data for BMC Public Health 2023
Weitere Titel:
-
Beschreibung:
(Abstract) Data for the study from Bollenbach, L., Niermann, C., Schmitz, J., & Kanning, M. 2023. Social Participation in the city. All descriptives and variables that were used for the calculation of the correlations, t-tests, and multigroup path analyses are included in this data set. The provided information is sourced from the article "Social participation in the city: exploring the moderating effect of walkability on the associations between active mobility, neighborhood perceptions, and social activities in urban adults" by Bollenbach, Niermann, Schmitz & Kanning, published in BMC Public Health, DOI: 10.1186/s12889-023-17366-0. All details are subject to the applicable availabilities and rights. The corresponding references can be found in the published article.
(Method) Study design: A cross-sectional online questionnaire was conducted and implemented via the German online platform ‘SoSci Survey’ [43]. The data were collected between July and December 2020 in the city of Stuttgart, Germany. For clarification, data collection took time during the COVID-19 pandemic, but no serious restrictions (e.g. curfews) were in place in the data collection timeframe. The first page of the questionnaire contained information about the study, its goals, data privacy protection, and participants’ rights in this context. To participate in the study, participants had to give informed consent that they were willing to participate and had read and understood the study information. However, this study includes only a part of all collected data (see ‘Measures’). The survey was in German.
(Method) Recruitment of the study participants: The individuals who participated in this study were recruited via the distribution of 3000 letters in 12 pre-selected residential areas in the city of Stuttgart, Germany. The letters contained information about the study background, a QR code to directly participate in the online questionnaire, and information about the option to participate via a paper-pencil questionnaire. Inclusion criteria were to live in the study area, to be at least 18 years old, and to understand German. The study sample is described in Table 1. The residential areas were pre-selected based on the objective walkability in the respective residential area, resulting in six residential areas with low walkability, and six residential areas with high walkability (see Fig. 1). The walkability scores for the classification of the pre-selected areas into low- and high walkability were derived via the first version of the ILS-Walkability-Index [35].
(Method) Measures: The following measures were derived and used in the data analyses to answer the research questions.
(Method) Social participation: To measure social participation, the scale used by Levasseur et al. [17] was adopted, which operationalizes social participation as participants’ frequency of monthly engagement in 10 different social activities. The response options to the question “How often are you involved in the following activities?” were rated on a 5-point Likert scale with the following indications: 1 (“never”), 2 (“less than once a month”), 3 (“at least once a month”), 4 (“at least once a week”), and 5 (“almost every day”). After data collection, the response options were converted into frequencies per month per activity (“almost every day” = 20; “at least once a week” = 6; “at least once a month” = 2; “less than once a month” = 1; and “never” = 0, respectively). In a final step, the frequencies from all 10 activities were summed, which resulted in the final social participation score that constitutes the number of social activities per individual per month with a theoretical range of 0-200 (note, it is hardly possible to be involved in every social activity on every day) [17]. One example for the question and a response option is as follows (see ‘Additional file 3’ for further information): ‘How often are you involved in the following activities?’ 1. Visit family members/friends.
(Method) Active mobility: To assess individuals’ level of active mobility, the validated ‘Physical Activity, Exercise, and Sport Questionnaire’ was used [44]. The questionnaire assesses various types of physical activities (such as everyday life activities, e.g., walking/cycling to work or leisure, household activities), exercises (for the purpose of physical activity itself, e.g., running, hiking), and sports (a more specific sport, often with a competitive character, e.g., soccer, track and field athletics). However, as for this study only walking and bicycling to work, for leisure, and for recreational purposes (active mobility) was of interest, only these measures were utilized. This resulted in a total of five items (1, walking to work; 2, walking to the grocery store; 3, bicycling to work; 4, bicycling for other transportation purposes; and 5, walking for recreation/strolling) that were summed up to the measure of active mobility. The items were assessed in the following manner [44]. After the introduction question “On how many days, and for how long have you conducted the following activities in the last four weeks?”, participants answered in cloze-type-questions, for example (see ‘Additional file 1’ for further information): Walking to work (also partial sections): On _ days during the 4 weeks and approximately _ minutes per day. With the information from the first (number of days of the respective activity) and the second (performed minutes per respective activity) response, the active mobility per month per participant (unit: minutes of active mobility per month per participant) was calculated and used in the analyses. This was done by summing up the products from each multiplication of days and minutes for 1, 2, 3, 4, and 5, respectively.
(Method) Neighborhood perceptions: Participants’ subjective neighborhood perceptions, i.e., their subjectively perceived satisfaction with the neighborhood environment, were measured via selected questions from the validated ‘Neighborhood Environment Walkability Scale - Germany’ (NEWS-G [4546]). To be precise, 10 questions from the subcategory ‘I’ (‘satisfaction with the neighborhood environment’) were assessed (see ‘Additional file 2’). The participants answered the questions regarding their satisfaction with different environmental features on a 5-point Likert scale, with answers ranging from 1 (“very unsatisfied”) to 5 (“very satisfied”). The final scale for analyses resulted from the mean of the answers. The scale had acceptable reliability with a Cronbach’s Alpha of 0.74. One example for a question and response is as follows (see ‘Additional file 2’ for further information): ‘How satisfied are you with…’… the possibility to walk in your neighborhood environment? Note that ‘subjectively perceived satisfaction with the neighborhood environment’ is abbreviated to ‘neighborhood perceptions’ in the rest of the manuscript to increase readability.
(Method) Walkability: First, the walkability measure that was used for the initial pre-selection of the 12 residential areas for participant recruitment was rechecked and updated with an adapted and improved version of the walkability measure that wasn’t available at the time of the initial data collection. We used the Walkability-Index from the Research Institute for Regional and Urban Development (= ‘Institut für Landes- und Stadtentwicklungsforschung’, ILS; ‘ILS-Walkability-Index’) to measure the objective walkability. The index was refined in the project ‘AMbit - Active Mobility’ [47] and is based on the basic concept of the original Walkability-Index, which was developed by Dobešová and Křivka [39]. We used new technical possibilities such as precise routing and open data [35]. Generation of the measure was done as follows: This objective walkability for the city of Stuttgart was determined using QGIS (a free and Open Source Geographic Information System software) to calculate the ILS-Walkability Index [35]. We calculated the walkability city-wide on a 500m by 500m grid and checked in which grid the participants live. For each cell of the grid, a score was calculated. The ILS-Walkability-Index consists of four dimensions: The permeability of the pedestrian network (data source: OpenStreetMap, European Digital Elevation Model), the proportion of green spaces (data source: OpenStreetMap), the population density (data source: German Zensus, 2011), and the availability of amenities (data source: OpenStreetMap) within walking distance. The permeability of the pedestrian network shows the area that a person can reach when walking 500m in any direction along the pedestrian network starting from the center of each cell. The result is a polygon – the so-called pedestrian shed. It is put in relation to the theoretical maximum size of the pedestrian shed – a circle with a radius of 500m. The higher the proportion is, the more permeable the pedestrian network is. An elevation model serves as a correction factor: The more meters of altitude, the smaller the pedestrian shed is. The proportion of green space is the proportion of the pedestrian shed that is covered with green space. Population density is derived from the number of residents living within the pedestrian shed. The accessibility of amenities is based on calculations of the distance along the walking network to different amenities such as supermarkets, schools, or restaurants. The closer and more numerous the amenities are, the higher the rating is. All four dimensions (permeability of the pedestrian network, green space, population density, amenities) are scaled from 0 to 10 and added together to the ILS-Walkability Score. Because population density correlates with the amenity-score (where many people live, there are a greater number of stores), a weight of 0.5 was applied to population density, while a weight of 1 was applied to the other dimensions. The sum is stretched to a scale from 0 to 50, where 50 represents the maximum walkability. Walkability was then categorized using tertiles. The first tertile includes grids with a range from 33 to 50 and corresponds to a high walkability. The second a range from 23 to 33 and corresponds to average walkability. Lower values correspond to a low walkability. Therefore, a value ≤ 23 indicates low walkability, and a value of ≥ 33 indicates high walkability. For the calculation, we used data from OpenStreetMap [48], the German Zensus 2011 [49], and the European Digital Elevation Model [50] to calculate the altitude. To calculate the distances, we used the OpenRouteService [51] from Heidelberg Institute for Geoinformation Technology [52]. For more details see [35].
(Method) Covariates: The demographics sex, age, height, weight, and socioeconomic status (SES) (Table 1.) were measured via self-report in the questionnaire. The SES represents a multidimensional index score that comprises three continuously measured components ‘Education and Occupational Qualifications’ (highest one achieved, e.g., Higher School Certificate), ‘Occupational Status’ (e.g., civil servant; comparative classification), and ‘Net Income’ that go into the index equivalently [53]. The SES had a possible range of 3–21 and was divided into 5 quintiles (low, 1. quintile, threshold = 6.6 (Q1); medium, 2.–4. quintile, threshold = 10.2 (Q2), 13.8 (Q3), 17.4 (Q4); high, (5. quintile, threshold = 21). Based on the self-reported height and weight, the BMI was calculated for each individual.
(Method) Study design: A cross-sectional online questionnaire was conducted and implemented via the German online platform ‘SoSci Survey’ [43]. The data were collected between July and December 2020 in the city of Stuttgart, Germany. For clarification, data collection took time during the COVID-19 pandemic, but no serious restrictions (e.g. curfews) were in place in the data collection timeframe. The first page of the questionnaire contained information about the study, its goals, data privacy protection, and participants’ rights in this context. To participate in the study, participants had to give informed consent that they were willing to participate and had read and understood the study information. However, this study includes only a part of all collected data (see ‘Measures’). The survey was in German.
(Method) Recruitment of the study participants: The individuals who participated in this study were recruited via the distribution of 3000 letters in 12 pre-selected residential areas in the city of Stuttgart, Germany. The letters contained information about the study background, a QR code to directly participate in the online questionnaire, and information about the option to participate via a paper-pencil questionnaire. Inclusion criteria were to live in the study area, to be at least 18 years old, and to understand German. The study sample is described in Table 1. The residential areas were pre-selected based on the objective walkability in the respective residential area, resulting in six residential areas with low walkability, and six residential areas with high walkability (see Fig. 1). The walkability scores for the classification of the pre-selected areas into low- and high walkability were derived via the first version of the ILS-Walkability-Index [35].
(Method) Measures: The following measures were derived and used in the data analyses to answer the research questions.
(Method) Social participation: To measure social participation, the scale used by Levasseur et al. [17] was adopted, which operationalizes social participation as participants’ frequency of monthly engagement in 10 different social activities. The response options to the question “How often are you involved in the following activities?” were rated on a 5-point Likert scale with the following indications: 1 (“never”), 2 (“less than once a month”), 3 (“at least once a month”), 4 (“at least once a week”), and 5 (“almost every day”). After data collection, the response options were converted into frequencies per month per activity (“almost every day” = 20; “at least once a week” = 6; “at least once a month” = 2; “less than once a month” = 1; and “never” = 0, respectively). In a final step, the frequencies from all 10 activities were summed, which resulted in the final social participation score that constitutes the number of social activities per individual per month with a theoretical range of 0-200 (note, it is hardly possible to be involved in every social activity on every day) [17]. One example for the question and a response option is as follows (see ‘Additional file 3’ for further information): ‘How often are you involved in the following activities?’ 1. Visit family members/friends.
(Method) Active mobility: To assess individuals’ level of active mobility, the validated ‘Physical Activity, Exercise, and Sport Questionnaire’ was used [44]. The questionnaire assesses various types of physical activities (such as everyday life activities, e.g., walking/cycling to work or leisure, household activities), exercises (for the purpose of physical activity itself, e.g., running, hiking), and sports (a more specific sport, often with a competitive character, e.g., soccer, track and field athletics). However, as for this study only walking and bicycling to work, for leisure, and for recreational purposes (active mobility) was of interest, only these measures were utilized. This resulted in a total of five items (1, walking to work; 2, walking to the grocery store; 3, bicycling to work; 4, bicycling for other transportation purposes; and 5, walking for recreation/strolling) that were summed up to the measure of active mobility. The items were assessed in the following manner [44]. After the introduction question “On how many days, and for how long have you conducted the following activities in the last four weeks?”, participants answered in cloze-type-questions, for example (see ‘Additional file 1’ for further information): Walking to work (also partial sections): On _ days during the 4 weeks and approximately _ minutes per day. With the information from the first (number of days of the respective activity) and the second (performed minutes per respective activity) response, the active mobility per month per participant (unit: minutes of active mobility per month per participant) was calculated and used in the analyses. This was done by summing up the products from each multiplication of days and minutes for 1, 2, 3, 4, and 5, respectively.
(Method) Neighborhood perceptions: Participants’ subjective neighborhood perceptions, i.e., their subjectively perceived satisfaction with the neighborhood environment, were measured via selected questions from the validated ‘Neighborhood Environment Walkability Scale - Germany’ (NEWS-G [4546]). To be precise, 10 questions from the subcategory ‘I’ (‘satisfaction with the neighborhood environment’) were assessed (see ‘Additional file 2’). The participants answered the questions regarding their satisfaction with different environmental features on a 5-point Likert scale, with answers ranging from 1 (“very unsatisfied”) to 5 (“very satisfied”). The final scale for analyses resulted from the mean of the answers. The scale had acceptable reliability with a Cronbach’s Alpha of 0.74. One example for a question and response is as follows (see ‘Additional file 2’ for further information): ‘How satisfied are you with…’… the possibility to walk in your neighborhood environment? Note that ‘subjectively perceived satisfaction with the neighborhood environment’ is abbreviated to ‘neighborhood perceptions’ in the rest of the manuscript to increase readability.
(Method) Walkability: First, the walkability measure that was used for the initial pre-selection of the 12 residential areas for participant recruitment was rechecked and updated with an adapted and improved version of the walkability measure that wasn’t available at the time of the initial data collection. We used the Walkability-Index from the Research Institute for Regional and Urban Development (= ‘Institut für Landes- und Stadtentwicklungsforschung’, ILS; ‘ILS-Walkability-Index’) to measure the objective walkability. The index was refined in the project ‘AMbit - Active Mobility’ [47] and is based on the basic concept of the original Walkability-Index, which was developed by Dobešová and Křivka [39]. We used new technical possibilities such as precise routing and open data [35]. Generation of the measure was done as follows: This objective walkability for the city of Stuttgart was determined using QGIS (a free and Open Source Geographic Information System software) to calculate the ILS-Walkability Index [35]. We calculated the walkability city-wide on a 500m by 500m grid and checked in which grid the participants live. For each cell of the grid, a score was calculated. The ILS-Walkability-Index consists of four dimensions: The permeability of the pedestrian network (data source: OpenStreetMap, European Digital Elevation Model), the proportion of green spaces (data source: OpenStreetMap), the population density (data source: German Zensus, 2011), and the availability of amenities (data source: OpenStreetMap) within walking distance. The permeability of the pedestrian network shows the area that a person can reach when walking 500m in any direction along the pedestrian network starting from the center of each cell. The result is a polygon – the so-called pedestrian shed. It is put in relation to the theoretical maximum size of the pedestrian shed – a circle with a radius of 500m. The higher the proportion is, the more permeable the pedestrian network is. An elevation model serves as a correction factor: The more meters of altitude, the smaller the pedestrian shed is. The proportion of green space is the proportion of the pedestrian shed that is covered with green space. Population density is derived from the number of residents living within the pedestrian shed. The accessibility of amenities is based on calculations of the distance along the walking network to different amenities such as supermarkets, schools, or restaurants. The closer and more numerous the amenities are, the higher the rating is. All four dimensions (permeability of the pedestrian network, green space, population density, amenities) are scaled from 0 to 10 and added together to the ILS-Walkability Score. Because population density correlates with the amenity-score (where many people live, there are a greater number of stores), a weight of 0.5 was applied to population density, while a weight of 1 was applied to the other dimensions. The sum is stretched to a scale from 0 to 50, where 50 represents the maximum walkability. Walkability was then categorized using tertiles. The first tertile includes grids with a range from 33 to 50 and corresponds to a high walkability. The second a range from 23 to 33 and corresponds to average walkability. Lower values correspond to a low walkability. Therefore, a value ≤ 23 indicates low walkability, and a value of ≥ 33 indicates high walkability. For the calculation, we used data from OpenStreetMap [48], the German Zensus 2011 [49], and the European Digital Elevation Model [50] to calculate the altitude. To calculate the distances, we used the OpenRouteService [51] from Heidelberg Institute for Geoinformation Technology [52]. For more details see [35].
(Method) Covariates: The demographics sex, age, height, weight, and socioeconomic status (SES) (Table 1.) were measured via self-report in the questionnaire. The SES represents a multidimensional index score that comprises three continuously measured components ‘Education and Occupational Qualifications’ (highest one achieved, e.g., Higher School Certificate), ‘Occupational Status’ (e.g., civil servant; comparative classification), and ‘Net Income’ that go into the index equivalently [53]. The SES had a possible range of 3–21 and was divided into 5 quintiles (low, 1. quintile, threshold = 6.6 (Q1); medium, 2.–4. quintile, threshold = 10.2 (Q2), 13.8 (Q3), 17.4 (Q4); high, (5. quintile, threshold = 21). Based on the self-reported height and weight, the BMI was calculated for each individual.
Schlagworte:
Urban environment
Social Participation
Active mobility
Subjective neighborhood perceptions
Walkability
Physical activity
Environmental perceptions
Social activities
City
Cross-sectional
Social Participation
Active mobility
Subjective neighborhood perceptions
Walkability
Physical activity
Environmental perceptions
Social activities
City
Cross-sectional
Sprache:
-
Herausgeber/in:
Universität Konstanz
Erstellungsjahr:
Fachgebiet:
Sports
Objekttyp:
(Dataset) cross-sectional data from an online questionnaire
Datenquelle:
-
Verwendete Software:
-
Datenverarbeitung:
-
Erscheinungsjahr:
Rechteinhaber/in:
Förderung:
Deutsche Forschungsgemeinschaft
(Grant 421868672)
Open Access funding enabled and organized by Project DEAL
Open Access funding enabled and organized by Project DEAL
Status:
Publiziert
Eingestellt von:
0491d3d609340a84715aad185b8882c2
Erstellt am:
Archivierungsdatum:
2023-12-01
Archivgröße:
39,4 kB
Archiversteller:
a73b8c277d10e12cb86f91edde66677a
Archiv-Prüfsumme:
eca5e95da9ff5d51722ebe518307d911
(MD5)
DOI: 10.48606/52
Publikationsdatum:
2023-12-01
Metadaten herunterladen
Statistik
0
Views
Views
0
Downloads
Downloads
Datenpaket zitieren
Bollenbach, Lukas
(2023): cross-sectional data for BMC Public Health 2023.
Universität Konstanz.
DOI: 10.48606/52