SAS 12: DENTAL SURVEY

importance of survey

as crucial as examinations in private health

essential for assessing oral health conditions in a community

involves asking individuals for information, usually via questionnaires

its success relies on its clearly defined purpose, rather than just on the respondents or their incentives

important to establish one primary purpose for the survey, supported by two to three SMART-form objectives

it is defined as an examination or appraisal to collect data for analyzing a group or area, according to Merriam-Webster

step 1 in planning a health survey

prepare a detailed written statement outlining the objectives of the health survey

clearly state the objectives and ensure they are feasible with available resources

consult local information to determine if existing data can be utilized for the proposed survey

step 2 in planning a health survey

specify definitions and classification criteria for the data

identify the items of information required for the health survey

define and include only methods of data collection based on survey objectives

elaborate on the use of each data item concerning classification, tabulation, and analysis

step 3 in planning a health survey

the ideal reference population is a group of healthy individuals

the sampling frame is a complete specification of the population to be sampled

reference population should be defined from which the information is to be sought

a clear definition aids in selecting the appropriate sampling procedure and interpreting data

due to the complexity of defining health, inclusion and exclusion criteria must be outlined based on the study's purpose

it is considered as a subset of the population that will serve as a standard against which the research findings are to be evaluated

step 4 in planning a health survey

deciding if the reference population should be included in the whole sample

consider the size of the reference population in relation to the resources available for the study

weigh the advantages and disadvantages of sampling versus conducting a comprehensive survey.

step 5 in planning a health survey

decide to study only a sample of the population

determine the number of units in the population for the survey

compute the optimum sample size, typically using Slovene’s formula

the optimum sample size is influenced by the prevalence or variability of the condition and the desired precision.

step 6 in planning a health survey

should be a fair representation of the population

focus on how respondents will be selected from the population making use of the sampling methods

step 7 in planning a health survey

part of structuring the survey

a good questionnaire must be tested and validated prior to survey implementation

new questionnaires require validation at least three times before acceptance for use

previously accepted questionnaires are considered standard and do not need further validation

the design and validation of the health survey questionnaire or observation forms are crucial for success

step 8 in planning a health survey

uniformity in performance is ensured through proper training

the selection and training of interviewers are crucial for a health survey

including trials or dummy runs is recommended as part of the training process

interviewers must be carefully chosen and trained in interviewing techniques

a thorough understanding of definitions, criteria, and methods is essential for data accuracy and reliability

step 9 in planning a health survey

preliminary editing of the collected data

testing and checking of equipment if used

this is the actual data gathering in the field

correct identification of selected sampling units

transportation arrangements for data collectors

retrieval of completed forms for initial processing

supervision and monitoring of interviewers during the process

publicity to inform and garner cooperation from the population

step 10 in planning a health survey

the data gathered is prepared to be analyzed.

the arrangements are needed for data analysis facilities to be available at hand.

step 11 in planning a health survey

the last phase is the writing of the report

population

refers to the entire group or in some books being referred to as the universe

sample

is the subset of the population or just a specific group out of the entire population.

2 general types of sampling methods

probability sampling methods

non-probability sampling methods

probability sampling methods

gives all individuals in the population to have a chance to be selected for the study

examples of probability sampling methods

simple random sampling

systematic sampling

stratified sampling

clustered sampling

multistage sampling

simple random sampling

gives every member of the population an equal chance of selection

techniques: random number generators & fish-bowl method

fish-bowl technique 

placing names or numbers in a container for random selection

advantages of simple random sampling

ease of calculating estimates

straightforward implementation

disadvantages of simple random sampling

impractical for small sampling units

challenges in achieving a complete sampling frame

potential underrepresentation of uncommon characteristics

contacting selected individuals can be inconvenient due to geographical dispersion

could exclude minority subgroups if they are underrepresented in the population

systematic sampling

the initial selection uses simple random sampling to determine which individuals are included

individuals are selected at regular intervals from the sampling frame to ensure an adequate sample size

involves selecting every nth unit from the population in the sampling frame, defined by a sampling fraction of 1/k

to obtain a sample size n from a population of size x, every x/nth individual is chosen; for instance, to sample 200 from 2000, select every 10th member

advantage of systematic sampling

the sample is evenly spread over the entire population

disadvantage of systematic sampling

the potential for bias if hidden periodicity aligns with the selection intervals

stratified sampling

dividing a population into subgroups or strata based on characteristics of interest

ensures representation from all subgroups, particularly when measurements may vary among them

for example, in a dental problems study, the population may be stratified by sex to ensure equal representation of men and women

advantage of stratified sampling

reduces sampling bias

enhances accuracy and representativeness of the sample

samples can be taken in equal sizes from each stratum, or non-equal sizes based on the proportions of the strata in the population

disadvantage of stratified sampling

challenge in determining which characteristics should be used for stratification

clustered sampling

two-stage sampling — preferred for large populations to reduce costs

divides a population into clusters of homogeneous units, typically based on geography

single-stage cluster sampling — all units from the selected cluster are included in the study

two-stage cluster sampling — individuals are randomly selected from each cluster for inclusion

advantage of clustered sampling

can be more efficient than simple random sampling over extensive geographical areas

disadvantage of clustered sampling

potential bias if selected clusters are not representative, which can increase sampling error

sampling error tends to be higher than in simple random sampling for the same sample size

multistage sampling

the process may use three or more stages

the selection is done in stages until the final sampling units are chosen.

a sample is selected at random with the probability of selection proportional to the size.

advantage:

• to cut the cost in doing the study.

disadvantage:

• the sampling error might be increased

non-probability sampling methods

involves non-random selection based on criteria allowing you to easily collect data.

examples of non-probability sampling methods

quota sampling

snowball sampling

convenience sampling

judgement (judgmental / purposive) sampling

convenience sampling

perhaps the easiest method of sampling

normally useful results can be obtained, but the results are prone to significant bias

participants are selected based on their availability and willingness to take part in the study

quota sampling

often used by market researchers

interviewers are given a quota of subjects of a specified type to attempt to recruit

ideally the quotas chosen would proportionally be always representing the characteristics, of the underlying population though not as always

judgement sampling

aka: judgmental / purposive / selective / subjective

relies on the judgement of the researcher in choosing who to ask to participate

researchers may implicitly choose a “representative” sample to suit their needs

specifically approach individuals with certain characteristics that they need to include in the study

advantage:

• time-and cost-effective to perform

snowball sampling

can be effective when a sampling frame is difficult to identify

used in social sciences when investigating for the hard-to-reach groups or the hidden population.

existing subjects identified are asked to nominate further subjects known to them to increases size like a rolling snowball

requirements of a good survey questions

objective

avoid sensitive ones

not use abbreviations

allow only one correct choice

not allow erroneous assumptions

use clear and easily understood terms

allow for multiple answers if necessary

allow to use neutral questions and answers

have a continuity with the previous question

give the respondent the choice not to answer

should include the most bothersome situation

use simple and clear questions and language easily understood

allow screening questions-same questions that will be asked in a different manner to check if the respondents are consistent