|
The National Citizen SurveyTM
<br />3 Suirve ,_sponse Rate
<br />Total sample used
<br />I=Complete Interviews
<br />P=Partial Interviews
<br />R=Refusal and break off
<br />NC=Non Contact
<br />0=Other
<br />1,600
<br />463
<br />2
<br />0
<br />0
<br />0
<br />UH=Unknown household 0
<br />UO=Unknown other 1,115
<br />Response rate: (I+P)/(I+P) + (R+NC+0) + (UH+UO) 29%
<br />,,,,,,,� ,,,IIIIII ��� III. »> �
<br />��yilly N'r(cfJJ���,, a +uu �ia�o ��/a�a INIU///�, /pmr /p� Ia�PIc/{/�f a wu �1�%'�u i lM a y� 0 !�t�mluol
<br />"llllq�„,; �fnnlll%'� IIIII IIIII �I IIIII "%CrpID . �"�,iroirrv��f IIIII II�II. p���///r nnm/ W�������m IIIII IIIII. ���ll�uv'�l�in��o IIIII) ���"JI� V�llii l
<br />Waduil!
<br />It is customary to describe the precision of estimates made from surveys by a "level of confidence" and
<br />accompanying "confidence interval" (or margin of error). A traditional level of confidence, and the one used here,
<br />is 95%. The 95% confidence interval can be any size and quantifies the sampling error or imprecision of the survey
<br />results because some residents' opinions are relied on to estimate all residents' opinions.
<br />The margin of error for the City of Ramsey survey is no greater than plus or minus five percentage points around
<br />any given percent reported for all respondents (465 completed surveys).
<br />For subgroups of responses, the margin of error increases because the number of respondents for the subgroup is
<br />smaller. For subgroups of approximately 10o respondents, the margin of error is plus or minus 10 percentage
<br />points.
<br />U ,Jli IIIIIhlllletl %e Ig a Vt
<br />Upon receipt, completed surveys were assigned a unique identification number. Additionally, each survey was
<br />reviewed and "cleaned" as necessary. For example, a question may have asked a respondent to pick two items out
<br />of a list of five, but the respondent checked three; in this case, NRC would use protocols to randomly choose two
<br />of the three selected items for inclusion in the dataset.
<br />All surveys then were entered twice into an electronic dataset; any discrepancies were resolved in comparison to
<br />the original survey form. Range checks as well as other forms of quality control were also performed.
<br />J; u��>I i���� IIIIII�rrrr�,��G// �/�,�,V(�i� y IIII 1� ��� ������� �� ���1 � % IIIII �GI
<br />�,��,� � �» ,, ��,�li �� III � �,�, I „�, d,,,� , � � ��� ;Ilea„�
<br />/ ,���� IIIIII I„rrr ,�,�
<br />1 �il,,,,,,, IIIII IIIII �ffit , f�
<br />The demographic characteristics of the survey respondents were compared to those found in the 2010 Census and
<br />American Community Survey estimates for adults in the City of Ramsey. The primary objective of weighting
<br />survey data is to make the survey respondents reflective of the larger population of the community. The
<br />characteristics used for weighting were sex and age. No adjustments were made for design effects. The results of
<br />the weighting scheme are presented in the following table.
<br />2 A 95% confidence interval indicates that for every 100 random samples of this many residents, 95 of the confidence intervals created will
<br />include the "true" population response. This theory is applied in practice to mean that the "true" perspective of the target population lies
<br />within the confidence interval created for a single survey. For example, if 75% of residents rate a service as "excellent" or "good," then the
<br />4% margin of error (for the 95% confidence interval) indicates that the range of likely responses for the entire community is between 71%
<br />and 79%. This source of uncertainty is called sampling error. In addition to sampling error, other sources of error may affect any survey,
<br />including the non -response of residents with opinions different from survey responders. Though standardized on The NCS, on other surveys,
<br />differences in question wording, order, translation and data entry, as examples, can lead to somewhat varying results.
<br />34
<br />
|