SocialCohesion

Neighbourhood Social Cohesion

Measuring: Perceptions of neighbourhood belonging, social cohesion, trust, safety, disorder, loneliness, fairness, cleanliness, and availability of local help within approximately a 20-minute walk or one kilometre of the participant’s home.

Number of Items: 9

Item	Question Wording	TILDA Variable
1	I really feel part of this area / I feel that I don’t belong in this area.	SCQNghBH1
2	Vandalism and graffiti are a big problem in this area / There is no problem with vandalism and graffiti in this area.	SCQNghBH2
3	I often feel lonely living in this area / I have never felt lonely living in this area.	SCQNghBH3
4	Most people in this area can be trusted / Most people in this area can’t be trusted.	SCQNghBH4
5	People would be afraid to walk alone after dark in this area / People feel safe walking alone after dark in this area.	SCQNghBH5
6	Most people in this area are friendly / Most people in this area are unfriendly.	SCQNghBH6
7	People in this area will take advantage of you / People in this area will always treat you fairly.	SCQNghBH7
8	This area is kept very clean / The area is always full of litter and rubbish.	SCQNghBH8
9	If you were in trouble, there are lots of people in this area who would help you / If you were in trouble, there is nobody in this area who would help you.	SCQNghBH9

Scoring Method:

Each item is answered using a 7-point bipolar response scale. The closer the participant’s response is to a statement, the more strongly they agree with that statement.

For the TILDA-derived neighbourhood dissatisfaction score (Neighb_unsatisfied):

Responses are coded from 1 to 7 from left to right.
Items 2, 3, 5 and 7 are reverse-scored using 8 - response so that all items are oriented in the same direction.
After alignment, responses 1–3 are coded 0 and responses 4–7 are coded 1.
The nine dichotomised items are summed to produce a score ranging from 0 to 9.
Higher scores indicate greater neighbourhood dissatisfaction / less favourable perceptions of the local area.
The TILDA derived score is calculated only when all nine component items are non-missing.

Note: The nine-item TILDA module covers several related neighbourhood constructs, including social cohesion, belonging, safety/disorder, loneliness and perceived support. The Neighb_unsatisfied summary is therefore a TILDA-derived composite rather than the standard four-item neighbourhood social cohesion score often used in ELSA/HRS research.

Citation:

Stafford, M., Bartley, M., Sacker, A., Marmot, M., Wilkinson, R., Boreham, R., & Thomas, R. (2003). Measuring the social environment: Social cohesion and material deprivation in English and Scottish neighbourhoods. Environment and Planning A, 35(8), 1459–1475. https://doi.org/10.1068/a35257
Cagney, K. A., Glass, T. A., Skarupski, K. A., Barnes, L. L., Schwartz, B. S., & Mendes de Leon, C. F. (2009). Neighborhood-level cohesion and disorder: Measurement and validation in two older adult urban populations. The Journals of Gerontology: Series B, 64B(3), 415–424. https://doi.org/10.1093/geronb/gbn041

Example TILDA Papers:

Nolan, A., O’Regan, C., Dooley, C., Wallace, D., Hever, A., Cronin, H., Hudson, E., & Kenny, R. A. (Eds.). (2014). The Over 50s in a Changing Ireland: Economic Circumstances, Health and Well-Being. The Irish Longitudinal Study on Ageing (TILDA), Trinity College Dublin. https://doi.org/10.38018/TildaRe.2014-00
Turner, N., Donoghue, O., & Kenny, R. A. (2018). Wellbeing and Health in Ireland’s Over 50s 2009–2016. The Irish Longitudinal Study on Ageing (TILDA), Trinity College Dublin. https://doi.org/10.38018/TildaRe.2018-00

Code

**Neighbourhood cohesion scale .. =>4 neg
foreach X in SCQNghBH1 SCQNghBH2 SCQNghBH3 SCQNghBH4 SCQNghBH5 SCQNghBH6 SCQNghBH7 SCQNghBH8 SCQNghBH9 {
        recode `X' (-99 = .) (-812 = 2) (-823 = 3) (-834 = 4) (-845 = 5) (-856 = 6) (-867 =7)
 }
recode SCQNghBH5 (8 = .)
 
 **Reverse code neg wording so all var are coded same direction
gen SCQNghBH2rev = 1 + 7 - SCQNghBH2
gen SCQNghBH3rev = 1 + 7 - SCQNghBH3
gen SCQNghBH5rev = 1 + 7 - SCQNghBH5
gen SCQNghBH7rev = 1 + 7 - SCQNghBH7

label var SCQNghBH2rev "There is no problem with vandalism and graffiti in this area"
label var SCQNghBH3rev "I have never felt lonely living in this area"
label var SCQNghBH5rev "People feel safe walking alone after dark in this area"
label var SCQNghBH7rev "People in this area will always treat you fairly"

**Neighbourhood dichotomised
foreach X in SCQNghBH1 SCQNghBH2rev SCQNghBH3rev SCQNghBH4 SCQNghBH5rev SCQNghBH6 SCQNghBH7rev SCQNghBH8 SCQNghBH9 {
        recode `X' (0/3 = 0) (4/max = 1)
        label def `X' 0 "Agree" 1 "Disagree", modify
        label val `X' `X'
}

**add together scale, higher the number higher the level of unsatisfaction!
gen Neighb_unsatisfied = SCQNghBH1 +SCQNghBH2rev +SCQNghBH3rev +SCQNghBH4 +SCQNghBH5rev +SCQNghBH6 +SCQNghBH7rev +SCQNghBH8 +SCQNghBH9

Download Stata .do file

ngh_items <- paste0("SCQNghBH", 1:9)

for (v in ngh_items) {
  x <- data[[v]]
  x[!is.na(x) & x == -99]  <- NA_real_
  x[!is.na(x) & x == -812] <- 2
  x[!is.na(x) & x == -823] <- 3
  x[!is.na(x) & x == -834] <- 4
  x[!is.na(x) & x == -845] <- 5
  x[!is.na(x) & x == -856] <- 6
  x[!is.na(x) & x == -867] <- 7
  data[[v]] <- x
}

# Item 5: value 8 is missing
data$SCQNghBH5[!is.na(data$SCQNghBH5) & data$SCQNghBH5 == 8] <- NA_real_


# ---------------------------------------------------------------
# Reverse negatively worded items
# Reverse formula: 8 - original response
# ---------------------------------------------------------------

data$SCQNghBH2rev <- 8 - data$SCQNghBH2
data$SCQNghBH3rev <- 8 - data$SCQNghBH3
data$SCQNghBH5rev <- 8 - data$SCQNghBH5
data$SCQNghBH7rev <- 8 - data$SCQNghBH7

attr(data$SCQNghBH2rev, "label") <-
  "There is no problem with vandalism and graffiti in this area"
attr(data$SCQNghBH3rev, "label") <-
  "I have never felt lonely living in this area"
attr(data$SCQNghBH5rev, "label") <-
  "People feel safe walking alone after dark in this area"
attr(data$SCQNghBH7rev, "label") <-
  "People in this area will always treat you fairly"


# ---------------------------------------------------------------
# Dichotomise neighbourhood items
# 0 = Agree
# 1 = Disagree
# ---------------------------------------------------------------

cohesion_items <- c(
  "SCQNghBH1",
  "SCQNghBH2rev",
  "SCQNghBH3rev",
  "SCQNghBH4",
  "SCQNghBH5rev",
  "SCQNghBH6",
  "SCQNghBH7rev",
  "SCQNghBH8",
  "SCQNghBH9"
)

for (v in cohesion_items) {
  x <- data[[v]]
  out <- rep(NA_real_, length(x))
  out[!is.na(x) & x >= 0 & x <= 3] <- 0
  out[!is.na(x) & x >= 4] <- 1
  data[[v]] <- out
  attr(data[[v]], "labels") <- c("Agree" = 0, "Disagree" = 1)
}


# ---------------------------------------------------------------
# Total scale
# Higher scores indicate greater neighbourhood dissatisfaction
# Ordinary addition preserves the Stata complete-case behaviour.
# ---------------------------------------------------------------

data$Neighb_unsatisfied <-
  data$SCQNghBH1 +
  data$SCQNghBH2rev +
  data$SCQNghBH3rev +
  data$SCQNghBH4 +
  data$SCQNghBH5rev +
  data$SCQNghBH6 +
  data$SCQNghBH7rev +
  data$SCQNghBH8 +
  data$SCQNghBH9

attr(data$Neighb_unsatisfied, "label") <-
  "Neighbourhood social cohesion dissatisfaction score"

Download R file

RECODE
  SCQNghBH1 SCQNghBH2 SCQNghBH3 SCQNghBH4 SCQNghBH5
  SCQNghBH6 SCQNghBH7 SCQNghBH8 SCQNghBH9
  (-99=SYSMIS)
  (-812=2)
  (-823=3)
  (-834=4)
  (-845=5)
  (-856=6)
  (-867=7).

RECODE SCQNghBH5 (8=SYSMIS).


* ---------------------------------------------------------------.
* Reverse negatively worded items.
* Reverse formula: 8 - original response.
* ---------------------------------------------------------------.

COMPUTE SCQNghBH2rev = 8 - SCQNghBH2.
COMPUTE SCQNghBH3rev = 8 - SCQNghBH3.
COMPUTE SCQNghBH5rev = 8 - SCQNghBH5.
COMPUTE SCQNghBH7rev = 8 - SCQNghBH7.

VARIABLE LABELS
  SCQNghBH2rev
    "There is no problem with vandalism and graffiti in this area"
  SCQNghBH3rev
    "I have never felt lonely living in this area"
  SCQNghBH5rev
    "People feel safe walking alone after dark in this area"
  SCQNghBH7rev
    "People in this area will always treat you fairly".


* ---------------------------------------------------------------.
* Dichotomise neighbourhood items.
* 0 = Agree; 1 = Disagree.
* ---------------------------------------------------------------.

RECODE
  SCQNghBH1 SCQNghBH2rev SCQNghBH3rev SCQNghBH4
  SCQNghBH5rev SCQNghBH6 SCQNghBH7rev SCQNghBH8 SCQNghBH9
  (0 THRU 3=0)
  (4 THRU HIGHEST=1).

VALUE LABELS
  SCQNghBH1 SCQNghBH2rev SCQNghBH3rev SCQNghBH4
  SCQNghBH5rev SCQNghBH6 SCQNghBH7rev SCQNghBH8 SCQNghBH9
  0 "Agree"
  1 "Disagree".


* ---------------------------------------------------------------.
* Total scale.
* Higher scores indicate greater neighbourhood dissatisfaction.
* Ordinary addition matches the Stata source: if any item is missing,
* Neighb_unsatisfied is missing.
* ---------------------------------------------------------------.

COMPUTE Neighb_unsatisfied =
  SCQNghBH1 +
  SCQNghBH2rev +
  SCQNghBH3rev +
  SCQNghBH4 +
  SCQNghBH5rev +
  SCQNghBH6 +
  SCQNghBH7rev +
  SCQNghBH8 +
  SCQNghBH9.

VARIABLE LABELS Neighb_unsatisfied
  "Neighbourhood social cohesion dissatisfaction score".

EXECUTE.

Download SPSS .sps file