Evaluating research methods

questionnaires / surveys

S: standardised questions increases reliability, less time consuming means larger pp can be used increasing generaliseability, informed consent improves validity

W: no explained meanings which can decrease validity, can have low response rates, decreasing generalisability, questions of sensative nature can cause people to give socially desireable answers = decrease validity

content analysis

S: is reliable as it can be carried out again, cheap method which means it can be repeated = reliable, no participants required = reliable

W: provides little information = lacking validity, time consuming means its not easily repeatable = reduced representativeness, analysis of text out of contect can decrease validity

structured interviews

S: reliable as method is numerical so similar results are likely, less time consuming meaning it can be more representative, informed consent is given = increased validity

W: not suitable with sensative issues as it can impact validity. Can be time consuming than other methods which reduces representativeness, can lead to people giving socially desirable answers which impacts validity

semi - structured interviews and unstructured

S: produces qualitative and valid data, flexibility of interviews can increase validity as in-depth insight can be gained, rapport can be built = higher validity

W: can become biased which decreases reliability, method is time consuming = decrease representativeness and generalisability, researcher bias impacts reliability

focus groups / group interviews

S: lots of in depth data = increased validity, pre-arranged = higher response rate and higher representativeness, rapport can be built, increasing validity

W: reliability is impacted as rapport can create bias, time consuming so small sample size decreases generalisability, validity is impacted if they give socially desireable answers

non participant overt observations

S: high validity as detailed notes are taken = getting a true insight, researcher is unlikely to be involved with the group, removing bias = high validity, informed consent is given = high validity

W: difficult to repeat = decreased reliability, risk of interpretation bias which decreases validity, subjects can feel emotional harm which reduces validity

non participant covert observations

S: observations are noted in a detailed way = improved validity, researcher isnt invovled with group so behaviour is genuine = validity, researcher is away from people if its dangerous = increased validity

W: reliability is reduced as it is difficult to replicate, high risk of interpretation bias which decreases validity, no informed consent means it is unethical and has low validity

participant observations overt

S: validity as observations are gathered in a detailed way, participants are likely to know about research aims, increasing representativeness, informed consent is given = increased validity

W: reliablility is low as it is difficult to repeat, high risk of interpretation bias, decreasing validity, may give socially desireable responses = decreased validty

participant observations covert

S: high validity as detailed notes are taken, can discretly probe respondants which increases validity, aware they’re in a study prevents respondants from coming to emotional harm due to informed consent given

W:difficult to replicate = decreased reliability, if researcher is involved its harder to take detailed notes = decreased validity, ethical issues as participants dont know they’re being observed.

ethnography

S: increased validity as researcher experiences lives of social group, more representative as researchers have better understanding of their study, researchers have informed consent from participants

W: not replicable = decreased reliability, small sample size decreases generalisability, can become less objective when immersed in the culture, reducing validity

official statistics

S: data can be replicated = high reliability, less time consuming as data is already collected = reliability, free from bias improves validity

W: research has low validity due lack of info behind statistics, lack of control over how data was collected = reduced validity, open to political abuse, meaning statistics may be manipulated decreasing validity

longitudinal study

S: evidence has social facts to reinforce which increases reliability, over time info is collected = true insight and high validity, rapport is built which increases validity

W:low in validity as it lacks little information to reinforce points, low generalisability as wider claims are harder to make, low validity as informed consent and protection from harm can change over time.