IGCSE Sociology: Research Methods and Inequality — Practice Questions

These are original questions written for Marlbridge, in the style and at the standard of the examination. They are not reproduced past-paper questions — examination boards hold copyright in their own papers. Use these alongside the official past papers available free from your board.

Related: Research Methods and Inequality revision notes

---

Section A

1. Define hypothesis, sample and pilot study. [3]

2. Name four sampling methods. [4]

Section B

3. Explain two advantages and two disadvantages of using structured interviews. [8]

4. Explain why a sociologist might carry out a pilot study before the main research, giving three reasons. [6]

5. Explain three explanations for the gender pay gap. [9]

6. Explain three ways in which ethnicity may affect a person’s life chances, and explain one difficulty in researching this area. [8]

Section C

7. Distinguish between reliability and validity as ways of evaluating a research method, and explain why quantitative methods tend to score higher on one and qualitative methods on the other. [6]

8. Explain what is meant by a sampling frame, and explain two ethical principles a sociologist must follow when carrying out research. [6]

9. Distinguish between ascribed status and achieved status, giving one example of each, and explain one piece of evidence used to support the view that identity is learned rather than natural. [6]

10. Explain how the “cycle of deprivation” can be used to explain the persistence of social class inequality across generations. [6]

---

Answers

1. Hypothesis — a testable statement predicting a relationship between two things, which the research sets out to support or reject [1]. Sample — the group actually studied, selected to represent the wider population [1]. Pilot study — a small-scale trial run of the research carried out before the main study [1].

2. Any four: random, systematic, stratified, quota, snowball, opportunity [1] [1] [1] [1].

3. Advantages: they are standardised — every respondent is asked the same questions in the same order, so the results are reliable and can be replicated and compared [1] [1]. The interviewer is present to clarify misunderstandings and to prompt, so response rates are much higher than for postal questionnaires and fewer questions go unanswered [1] [1]. Disadvantages: the fixed questions and closed answers give respondents no chance to explain or raise anything unanticipated, so the data lacks depth and validity [1] [1]. The presence of the interviewer may produce interviewer bias or socially desirable answers, particularly on sensitive topics, so respondents may not answer truthfully [1] [1]. They are also time-consuming and expensive compared with questionnaires, limiting sample size [1] [1].

4. Any three, 2 marks each: to check that the questions are clear and unambiguous, so that respondents interpret them as the researcher intends [1] [1]. To estimate how long the research will take and whether the method is practical, allowing the design to be adjusted before money is committed [1] [1]. To identify unforeseen problems, such as questions people refuse to answer or response options that do not fit anyone’s actual situation [1] [1]. To train interviewers and standardise their approach before the main study [1] [1].

5. Any three, 3 marks each: Occupational segregation — women are concentrated in sectors such as care, retail and clerical work, which are historically undervalued and low-paid [1] [1]; this is described as horizontal segregation, while vertical segregation means women are under-represented in senior posts within any given sector [1]. The motherhood penalty and the dual burden — women are far more likely to take career breaks and to work part-time to care for children [1] [1]; this interrupts promotion and pension accumulation, and part-time work is generally lower-paid per hour [1]. Discrimination and the glass ceiling — women may be passed over for promotion because of assumptions about commitment or availability [1] [1]; informal male networks and recruitment by similarity make senior roles harder to reach, and legislation has not eliminated this [1]. Socialisation and subject choice — girls and boys are channelled towards different subjects and career expectations from an early age, so the pay gap is partly established before employment begins [1] [1] [1].

6. Any three, 2 marks each: education — some minority ethnic groups experience lower attainment linked to teacher labelling, an ethnocentric curriculum and material deprivation, though attainment varies widely between groups [1] [1]. Employment — higher rates of unemployment and concentration in low-paid work, partly through discrimination at the recruitment stage, which field experiments using matched CVs have demonstrated [1] [1]. Housing — greater likelihood of living in poorer-quality or overcrowded housing in deprived areas, affecting health and educational opportunity [1] [1]. Health and policing — differences in health outcomes and in the likelihood of being stopped and searched [1] [1]. Difficulty in researching this area: ethnicity is a broad category that conceals enormous variation between and within groups, so aggregated statistics can mislead [1]; it is also difficult to separate the effect of ethnicity from that of social class, since the two overlap, and respondents may be reluctant to discuss discrimination with a researcher from a different background [1].

7. Reliability means that if the research were repeated using the same method it would produce the same result [1]; validity means the method gives a true and accurate picture of what is actually being studied [1]. Quantitative methods such as questionnaires and structured interviews use standardised, fixed-choice questions that are easy to repeat and compare, so they tend to be strong on reliability [1] [1]; but the fixed answers may not capture what a respondent actually means, so they can be weaker on validity [1]. Qualitative methods such as unstructured interviews and participant observation produce rich, detailed data that captures meaning accurately, so they tend to be strong on validity, but that same richness is harder to reproduce consistently, so they are typically weaker on reliability [1] [1].

8. A sampling frame is the list from which a sample is drawn; where no complete list of the target population exists, the researcher is often forced into a non-random sampling method, such as snowball or opportunity sampling [1] [1]. Any two, 2 marks each: informed consent — participants must understand what the research involves and agree to take part [1] [1]; confidentiality — participants’ identities and data must be protected [1] [1]; protection from harm — the research must not cause physical or psychological harm [1] [1]; the right to withdraw — participants may leave the study at any point [1] [1]. (Accept: covert observation breaches informed consent through deception, and is normally only defended where overt research would be impossible to carry out at all [1] [1].)

9. Ascribed status is fixed at birth and not earned, e.g. sex or the family a person is born into [1] [1]; achieved status is earned through a person’s own action, e.g. an occupation or qualification [1] [1]. Any one, 2 marks: cross-cultural variation — behaviour and norms assumed to be natural vary considerably between different societies, suggesting they are learned rather than innate [1] [1]; or evidence from feral children — children raised with little or no human contact fail to develop expected social behaviours, suggesting these behaviours are not automatic but require socialisation [1] [1].

10. The cycle of deprivation explains how disadvantage is passed from one generation to the next through a linked chain of causes: poor housing and diet contribute to poor health and educational underachievement [1] [1], which in turn lead to low-paid or insecure work in adulthood [1] [1], which reproduces poverty and disadvantage into the next generation, restarting the cycle [1] [1].

---

Where marks are usually lost

• Defining a hypothesis as a question rather than a testable statement.
• Giving advantages of interviews generally rather than of structured interviews.
• Explaining the gender pay gap solely through discrimination.
• Treating all minority ethnic groups as having identical outcomes.